We are currently working with a leading textile business to evolve and modernize their legacy ERP system through a custom implementation built on ERPNext and Frappe.
The project began with extensive business process discovery, reverse-engineering years of operational workflows, permissions, reports, and business rules from the existing system. By translating these legacy processes into structured requirements and system designs, we are helping the client preserve operational knowledge while moving to a modern, maintainable platform.
Our current focus is the Supply Chain module, where we have implemented textile-specific contract management, approval workflows, custom validations, reporting systems, and business process automation. Future phases will extend the platform into Administration, Accounts, and Griege management, creating a unified ERP tailored specifically for textile manufacturing operations.
Performative Café is a story-driven mobile cooking game inspired by the fast-paced chaos of cooperative kitchen simulators, blending time-management gameplay with a light mystery narrative.
Players step into the role of a young fry cook who got laid off, and gets his own cafe through a mysterious investor named the \"Algorithm\". Players face challenges presented by the Algorithm in increasing difficulties, to master the art of crafting picture-perfect food under pressure.
We took on this entire project from idea to complete delivery, delivering one of the first in the cooking game genre to be built in a portrait, mobile-first experience. The game features an explorable map, progression-based challenges, and an order system where players must prepare both hot and cold recipes accurately and efficiently. Performance is rewarded through a followers-based scoring system.
As the story unfolds, players will uncover the truth behind the challenges and the identity of the mysterious benefactor—revealing a personal connection that ties the café's journey together and brings the narrative full circle.
The game is currently live on the Oasiz platform, and is used for the training of Oasiz's AI game creation pipelines.
Banana of Doom is a fast-paced multiplayer card battler where players construct absurd sentences from word cards and submit them to an AI judge that decides the winner based on humor, context, and creativity.
We delivered a significant portion of the game—covering roughly 30–40% of both backend and frontend systems. This included core gameplay systems, AI provider orchestration (text and voice), and the infrastructure required to support real-time, judgment-driven matches at scale.
Beyond the core loop, we built extensive production features including beautiful animations, Steam Deck and desktop-native deployment, Discord-integrated activity flows, and a fully functional in-game shop with trading card game mechanics, rarities, and progression systems.
The result is a tightly engineered, high-velocity party experience where language, randomness, and AI judgment collide—powered by a robust multi-provider AI backend and designed for endless replayability.
Chess Revolution transforms competitive chess into a persistent progression experience, where every match contributes to the growth of a player's evolving island. We led the entire project from a single idea—\"We need a chess game in Oasiz\". Our work involved game design, art and engineering to create a complete multiplayer experience.
Players can challenge friends, compete against others, or practice with AI bots while climbing a fully integrated Elo ranking system. As players improve, their islands evolve to reflect their achievements.
We reimagined the board with \"living pieces\" inspired by the Tiny Swords pixel-art characters by Pixel Frog, allowing players to switch between classic chess pieces and animated warriors. On the technical side, we integrated Stockfish directly into the game, enabling fast local AI analysis without relying on external APIs.
The result is a fresh take on chess—combining timeless strategy with progression, personality, and modern game design.
The game is currently live on the Oasiz platform, and is used for the training of Oasiz's AI game creation pipelines.
Pro Pocket is our take on the classic 8 Ball Pool experience, built from the ground up for the Oasiz platform as a polished, mobile-first multiplayer game.
We designed Pro Pocket to deliver quick, competitive matches that are easy to jump into while maintaining the depth and strategy players expect from cue sports. Players can challenge friends in turn-based real-time matches or sharpen their skills against AI opponents across multiple difficulty levels.
The game features a complete multiplayer experience, including player invites, private match lobbies, table selection, aiming assistance, ball assignment systems, foul detection and handling, turn indicators, match progression, and detailed win/loss states.
To complement the gameplay and Oasiz's \"Club Jibbles\" theme, we built a distinctive pixel-art lounge environment that gives the game its own identity while reinforcing its social atmosphere. Combined with streamlined matchmaking and accessible controls, Pro Pocket transforms a familiar pool format into an engaging multiplayer experience designed for friendly competition, replayability, and social play on Oasiz.
The game is currently live on the Oasiz platform, and is used for the training of Oasiz's AI game creation pipelines.
In partnership with Semolina, we designed and developed a modern digital experience for one of Pakistan's most distinctive pasta brands. Built around the restaurant's identity, the website combines warm storytelling, product discovery, and e-commerce functionality into a cohesive customer journey.
The site showcases Semolina's handcrafted approach to pasta, family-run origins, and growing product lineup through a clean, responsive design optimized for both desktop and mobile users. We focused on creating an experience that feels as premium and approachable as the brand itself.
From visual design, art and user experience to frontend development and online ordering flows, we helped bring Semolina's story to life on the web—creating a platform that supports both its restaurant presence and expanding retail offerings.
Playroom had been quietly in the works for over two years with a simple goal: empowering the next generation of multiplayer games.
Recently, when the team saw strong adoption from AI-native companies, new founders and builders embracing vibe coding and Playroom gaining momentum there. Playroom Kit which already powered 25M+ players last year, was battle-tested to work at scale.
Grayhat helped them fully - from updates in their multiplayer SDK, to a design revamp for their landing, and a new AI-powered documentation site, to make it work beyond just games - with Playroom Kit, you can now one-shot build collaborative apps, multiplayer AI tools, and games from a single prompt, with zero server setup. Whatever vibe coding platform you use - Lovable, Base44, Replit, Cursor or more - Playroom is there to be your multiplayer infra that works at scale!
We also made the docs AI-native:
Zuno is a real-time multiplayer card game inspired by the timeless Crazy Eights formula, reimagined for the Oasiz platform with a focus on speed, accessibility, and competitive social multiplayer gameplay, in a Tiki Island theme to suit the \"Oasiz\" brand.
Built from the ground up in just two weeks, we delivered the complete experience—from game design and gameplay systems to visual assets, progression flow, and multiplayer implementation. The project covered every aspect of production, including card mechanics, turn management, match flow, user experience, and platform integration.
The game is currently live on the Oasiz platform, and is used for the training of Oasiz's AI game creation pipelines.
In a confidential enterprise engagement within a highly regulated industry, our Grayhatter worked on a large-scale cloud platform composed of dozens of interconnected services across a complex AWS microservices architecture. The project demanded the ability to quickly understand unfamiliar technologies, navigate distributed systems, and contribute reliably within a high-stakes engineering environment.
Over the course of the engagement, our engineer consistently exceeded expectations, earning repeated praise from the client for technical execution, ownership, and professionalism. They tackled challenging backend and infrastructure problems, improved scalability and maintainability through architectural refactors, contributed to cloud migration initiatives, and built automation workflows that reduced previously manual operational processes from hours to minutes.
Beyond technical delivery, the project became a testament to adaptability and engineering maturity. Working alongside senior international teams, our Grayhatter continuously expanded their expertise across new languages, frameworks, and cloud services while maintaining excellent communication, reliability, and a strong sense of ownership throughout the entire software development lifecycle.
We built Space Force as a fast-paced multiplayer space combat experience where players pilot agile starfighters through asteroid-filled arenas, battling in real-time matches for control of the leaderboard.
Designed around arcade-style movement and responsive controls, the game challenges players to navigate dynamic battlefields, collect power-ups, avoid hazards, and eliminate opponents to earn points. Every match rewards precision, positioning, and quick decision-making as players compete to dominate the arena before time runs out.
To support both competitive and casual playstyles, we included local and online multiplayer modes, AI opponents for practice, customizable match setups, multiple player slots, and a guided onboarding experience that introduces core mechanics through an integrated tutorial system. Live scoreboards, trophy-based progression, and social features encourage players to continuously improve, compete with friends, and climb the rankings.
The game is currently live on the Oasiz platform, and is used for the training of Oasiz's AI game creation pipelines.
Ruby is an AI-powered sales intelligence platform designed to help revenue teams discover opportunities, analyze accounts, and accelerate deal creation through intelligent workflows. Built around a modern account-centric experience, Ruby combines AI analysis, prospect research, and CRM integrations to help teams identify and prioritize their highest-value opportunities.
Working alongside the Ruby team, we contributed across the platform's core experience—from onboarding and account management to AI-powered account analysis, deal journeys, and the Analyst assistant. Users can import accounts from existing systems, run AI-driven analysis, communicate with Analyst using account-specific context, and convert AI-discovered opportunities into actionable sales deals.
We also helped expand Ruby's ecosystem through deep integrations with HubSpot, Google Calendar, Outlook Calendar, and billing infrastructure. These integrations enable teams to synchronize customer data, automate workflows, manage subscriptions, and maintain a unified view of their sales operations across multiple platforms.
Beyond feature development, we contributed extensive improvements to platform usability, including responsive design enhancements, settings management, onboarding flows, account management experiences, and UI refinements. The result is a comprehensive AI-native sales platform built to support the next generation of revenue teams.
PlayroomKit for Unity started as a straightforward idea for us: build a native C# bridge over an existing web SDK and make multiplayer feel effortless inside Unity. From that starting point, we scaled it from zero into a fully production-ready system used in hundreds of shipped games, including Little Umbrella’s biggest titles like \"Banana of Doom\".
We extended the core runtime with Unity Editor panels to improve developer experience and reduce integration friction, making setup and iteration significantly faster for teams building real-time games.
Over more than two years, we owned the entire lifecycle of the project—open source maintenance, documentation, bug triage, DevRel, support tickets, and major re-architectures—continuously evolving it to meet production demands and keep it stable at scale.
As part of the ongoing evolution of PlayroomKit, we were brought in to deliver a series of highly requested improvements to the popular JavaScript SDK and developer experience.
Our work focused on enhancing multiplayer reliability, player lifecycle management, and framework compatibility. We introduced support for modern React versions, implemented new room-leaving capabilities (the leaveRoom function), improved player reconnection and state preservation flows, refined joystick input handling, and contributed supporting documentation to help developers adopt the new features quickly.
While a relatively short engagement, the project directly impacted the day-to-day experience of PlayroomKit developers within Little Umbrella and outside of it, strengthening the SDK's foundation and helping make multiplayer game development more predictable, resilient, and enjoyable.
A complete digital rebuild for Rogers Marketing Group, focused on restoring structure, updating core settings, and delivering a full WordPress redesign from the ground up. The new site reframes the brand with a cleaner interface and stronger information hierarchy, better reflecting its strategic, results-driven positioning.
Built on WordPress, the project combined technical recovery work with modern UX and visual refinement—migrating, reconfiguring, and redeveloping the platform to improve performance, maintainability, and content clarity.
We created a refined digital presence for Guardian Alpha, in close collaboration with the client, which was designed to bring clarity and confidence to its healthcare offering. The website reimagines the brand through a custom-built visual system tailored for modern medical audiences, balancing precision with accessibility.
Built from scratch on WordPress, the project focuses on a fully custom theme design that prioritizes structure, usability, and trust.
Mindchat is an AI companion platform designed to create meaningful, emotionally engaging conversations through personalized AI friends. We focused on delivering an experience that feels less like a chatbot and more like a trusted companion, by researching with actual psychologists and counselors, then combining conversational AI, mood tracking, and immersive interactions into a mobile-first experience tailored for Gen Z users.
We researched, designed and developed the core product experience, including onboarding, authentication, mood tracking, AI chat, custom AI friends, voice calls, video interactions, and offline experiences. Users can create and interact with unique AI companions, track their emotional wellbeing over time, and engage through text, voice, and animated video conversations that feel natural and responsive.
A key part of the experience was building personalized AI avatars capable of realistic speech-driven animation. Instead of relying on simple looping animations, we developed characters whose facial movements and lip synchronization respond dynamically to what they are saying, using speech-aware visemes and real-time animation systems to create more expressive and believable conversations. Combined with custom personality systems, voice interactions, and conversational memory, these avatars helped transform AI companions into engaging digital characters rather than traditional chat interfaces.
Echelon AI Scriptwriter is a Notion-integrated AI tool that we built to transform video content into structured, production-ready scripts in seconds. By analyzing a video's content, context, and key talking points, we enable creators, marketers, and content teams to quickly generate high-quality scripts directly within their existing workflow. We also built support for manual script generation and refinement, allowing users to tailor outputs to their specific style and requirements. Additionally, users can interact with Gemini through custom prompts to iterate on, improve, and personalize scripts, creating a seamless AI-assisted content creation experience inside Notion.
Ares was an ambitious attempt by a founder in SF, to recreate the next generation of desktop assistants, inspired by emerging AI-native experiences like Sky.app. In an incredibly aggressive 2-week timeline, we approached the product not as just another AI chat application, but as a true Computer Use Agent capable of understanding and acting within a user's desktop environment.
Before writing a single feature, our team conducted extensive research into the evolution of desktop productivity — studying Spotlight, Alfred, ChatGPT Desktop, Claude Code, Manus, and Sky.app itself. We reverse-engineered user experiences, mapped workflows, and established a structured product development process to maximize velocity without sacrificing engineering discipline.
The resulting prototype featured a native-feeling Electron application with a Liquid Glass-inspired interface, achieving over 60% visual similarity to the original Sky experience. Under the hood, we integrated MCP servers and AppleScript-based desktop automation, allowing the AI to interact with macOS applications, execute multi-step workflows in a single request, create notes, manage files, and request explicit user permissions for privileged actions.
The project also explored a modern AI architecture using bundled MCP servers, local tool execution, and a flexible agent framework capable of expanding beyond simple chat interactions.
Although the project was ultimately shelved before completion, it demonstrated our ability to rapidly research, prototype, and engineer cutting-edge AI experiences at the frontier of human-computer interaction.
ARKA Creative is a storytelling-driven creative agency specializing in video production, animation, 3D experiences, and digital content. After collaborating with over 30 leading brands across Pakistan, ARKA expanded to the UK with a vision of bringing its cinematic storytelling and creative excellence to an international audience.
We partnered with ARKA to translate their existing creative vision into a high-performance digital experience. Built entirely in Framer, the website combines smooth animations, fully responsive layouts, SEO optimization, and a CMS-powered content system that allows their team to easily maintain and showcase their evolving portfolio.
The challenge was execution without compromise—preserving ARKA's bold identity, visual language, and narrative-first approach while ensuring the website matched the quality of the stories they create.
The result is a modern creative showcase designed to connect ARKA with international studios, brands, and businesses seeking powerful visual storytelling, animation, and next-generation digital experiences.
Upstreet, a Silicon Valley startup, is focused on the future of agentic AI. We helped them to help develop Upstreet SDK (USDK), a TypeScript framework that enables developers to create, manage, and deploy autonomous AI agents using familiar React-style development patterns.
Our Grayhat team members' work focused on building a developer experience that makes agent creation as intuitive as modern web development. We squashed many bugs, and helped the product reach its first public launch and open-source launch, and ensured zero happy-flow crashes on launch.
Agents built with USDK can interact with users, collaborate with other agents, and connect to external applications, APIs, and services. Whether automating workflows, managing communities, generating content, or performing complex multi-step tasks, developers can extend agents with tools that enable real-world interactions across existing systems using integrations. Rather than treating agents as simple workflows or chatbots, USDK allows developers to create persistent digital characters with their own personalities, appearances, memories, and unique speaking styles.
Alongside the SDK, we also developed and maintained the documentation platform, creating a comprehensive developer resource that helps users quickly learn, build, and deploy production-ready AI agents.
Enrichly is a discovery platform that connects parents with camps, activities, sports programs, and educational experiences through a single, intuitive marketplace. Designed to simplify what is often a fragmented search process, the platform helps families find opportunities tailored to their children's interests, age groups, and locations.
We spearheaded the product strategy, UI/UX design, and platform development, starting from just an idea, then into a proof-of-concept, a prototype, and finally into a complex multi-sided marketplace MVP. We designed an experience that felt approachable for both families and organizers, similar to Pinterest. The platform includes advanced search and filtering, interactive maps, child profile management, organization management, favorites, and role-based experiences for parents and activity providers.
Enrichly supports a wide range of categories—from STEM camps and arts programs to sports leagues and wellness activities—while providing organizers with the tools needed to establish their presence and manage their offerings.
The result is a modern marketplace that makes discovering enriching experiences easier for families while helping organizations reach the communities they serve.
Jawla is a multilingual trivia platform designed specifically for Middle Eastern audiences, bringing culturally relevant knowledge, entertainment, and competition together in a seamless web and iOS experience. We built an interactive Arabic-first game system featuring multiple game modes, category-based challenges, real-time scoring, rewards, and branded experiences—creating an engaging and accessible trivia platform tailored to regional players.
Working alongside Upstreet, we designed and developed a browser-based platform for creating, deploying, and managing AI agents without requiring developers to leave their workflow. Inspired by the simplicity of modern cloud platforms, the product provided a unified environment where teams could build agents, monitor deployments, configure integrations, and orchestrate complex agent ecosystems entirely from the browser.
Starting from a blank canvas, our Grayhatters helped shape the platform from initial product concept and UX direction through frontend and backend implementation, while also contributing to branding and launch materials. Users could create and configure both Mastra and ElizaOS agents directly within the platform, interact with them through integrated chat experiences, inspect logs and execution traces in real time, monitor container status, manage environment configuration, and deploy agents from reusable templates with a single click. The platform also supported collaborative multi-agent workflows, allowing specialized agents to work together on larger tasks and objectives.
A major focus of the platform was extensibility. Users could connect self-hosted MCP (Model Context Protocol) servers, enabling agents to securely access external tools, APIs, resources, and workflows through a standardized interface. The platform included a dedicated MCP registry, making it possible to connect hundreds of third-party services—including communication platforms, developer tools, and business systems—in just a few clicks. Combined with extensive observability, deployment controls, and agent orchestration capabilities, the result was a complete developer platform for building, deploying, and scaling production-ready AI agents.
Working with an entrepreneur in Kuwait, we helped bring a unique idea to life with PlayClawzy — a real-world claw machine connected to the internet through custom hardware APIs. The challenge was not just building a website, but creating a near real-time gaming experience where players could remotely see, control, and interact with a physical machine from anywhere.
Our Grayhat team designed and developed the frontend and backend experience around this hardware integration, collaborating closely with hardware engineers to bridge the gap between the physical and digital worlds. Players could join a queue, enter a live game session when their turn arrived, control the claw through intuitive on-screen controls, and claim prizes that could then be shipped directly to them.
Beyond the core gameplay loop, we expanded the platform with user profiles, operational monitoring tools, bug reporting systems, and additional social-first concepts including a TikTok-integrated experience that allowed users to discover and jump directly into the game.
We developed the official marketing landing page for Little Umbrella, an independent studio building AI-native multiplayer party games.
Little Umbrella is best known for experimental, social-first titles such as Death by AI, a breakout party game where players are placed in absurd, AI-generated survival scenarios and judged on their responses, as well as Banana of Doom, a fast-paced multiplayer party game built around chaotic, humor-driven interaction, and other AI-assisted social game concepts developed through their PlayroomKit ecosystem.
After the studio attracted backing from investors and operators across gaming and tech, including participation from Workplay Ventures (Mark Pincus’s venture firm) and other early-stage backers, we helped them bring their landing page to life.
For PickMeUp, a ride-sharing platform in Nigeria focused on connecting drivers and passengers, we played a key role in transforming a static and difficult-to-maintain frontend into a more scalable, content-driven experience. The platform was migrated from hardcoded content to a Strapi-powered CMS, enabling the client to manage website information, updates, and content without requiring developer involvement.
Beyond the architectural improvements, the project included the development of new sections such as FAQs and informational pages, alongside critical fixes to the driver and passenger onboarding flows. We refactored existing code, improved maintainability, and addressed issues that directly impacted user experience.
The contribution extended through the final stages of the product lifecycle, assisting with deployment and ensuring that the improved website was successfully prepared for production release.
IslamQ was envisioned as a first-of-its-kind Islamic knowledge platform in collaboration with the University of Madinah, designed to help users discover authentic answers from trusted scholars and sources through modern AI-powered technology.
We explored the intersection of artificial intelligence, semantic search, and religious knowledge management by designing a system capable of automatically scraping, processing, categorizing, translating, and indexing Islamic content across multiple online sources. The platform was architected with a rigorous human quality assurance workflow, ensuring scholars and volunteers could review, edit, and approve information before it reached end users.
Our work included a Flutter mobile application, a Python-powered data processing pipeline, automated web scrapers, multilingual English-Arabic content handling, and a semantic search engine built using vector databases and AI techniques to retrieve answers based on meaning rather than exact keywords. The system also considered accessibility features including speech-to-text, text-to-speech, and sign-language integrations to make Islamic knowledge available to a broader audience.
Prime is a premium mobility platform designed to bring transportation, convenience, and lifestyle services together in a single experience. Built for the UK market, the app connects users with trusted drivers, premium services, and a seamless booking experience through an elegant, mobile-first interface.
Working closely with the Prime team, we helped bring the product vision to life—product design, UI/UX, backend engineering and frontend (web and cross-platform) app development—across multiple stakeholder through polished mobile experiences focused on usability, speed, and trust. The platform was designed to feel less like a traditional ride-hailing application and more like a comprehensive urban lifestyle companion.
Beyond transportation, Prime embraces the concept of a superapp—creating a foundation for integrating multiple services into a unified ecosystem. From booking journeys to accessing curated experiences and partner services, the product is built around convenience and premium customer experiences.
The result is a sophisticated mobility platform that combines modern product design with intuitive user flows, helping users navigate London's fast-paced lifestyle through a single connected experience.
WEEPRO Coders ambitiously set out to transform its fragmented online education operation into a seamless digital ecosystem for young programmers. With students, parents, teachers, and administrators spread across multiple platforms like Zoom, WhatsApp, Google Drive, and coding environments, the challenge was to create a scalable workflow that could support rapid growth without increasing operational complexity.
We worked closely with the team through a discovery and product strategy phase, mapping the entire learner journey—from booking a free trial and onboarding parents to managing camp schedules, communication, attendance, and ongoing student progress. Rather than replacing familiar tools, we designed an integrated ecosystem centered around the way users already behaved.
Our solution leveraged Google Apps Script, Zoom, Moodle, calendar systems, and automated WhatsApp workflows to transform a highly manual process into a scalable operation. Automated trial booking, personalized reminders, registration confirmations, camp notifications, and parent communication reduced the need for hundreds of manually managed WhatsApp interactions.
Alongside the operational automation, we designed a future-facing learning platform with role-based experiences for learners, parents, teachers, and administrators, incorporating virtual classrooms, progress tracking, resource management, and analytics.
Herewith is a healthcare platform that connects older adults with trusted caregivers, making it easier to discover, book, and manage personalized care services. Working alongside Procyon Technostructure’s distributed engineering team across the United States and India, Grayhat contributed to the development of Herewith’s web platform.
Our Grayhatter joined the project at a critical stage, taking ownership of complex Next.js frontend flows and rapidly becoming a leading force within the web engineering effort. From implementing large product areas such as shift tracking, notifications, offers, disputes, and other caregiver workflows to navigating changing requirements and tight delivery timelines, they helped transform a struggling codebase into a product ready for customer release.
Beyond writing code, the engagement required technical leadership, realistic planning, and close collaboration with cross-functional teams across multiple time zones. The resulting web application successfully reached release, with the web team receiving direct appreciation from the client for its turnaround, execution, and commitment under pressure.
Dorbi set out to implement its first e-commerce venture in the UAE, combining its decades of expertise in auto accessories, detailing, and metal finishing into a modern digital storefront. As the exclusive Middle East distributor of AlloyGator wheel protectors, their flagship product, Dorbi needed an experience that reflected the premium nature of the vehicles and products they serve.
We designed and engineered a sleek, race-inspired platform that transformed traditional online shopping into an immersive automotive experience. Built with Next.js and Medusa, the platform included product discovery, customer support workflows, and a unique \"Try the Product\" experience where customers could visualize and customize accessories directly on a digital vehicle.
Every aspect of the experience was crafted with the language of motorsports in mind—bold visuals, aggressive styling, high-performance interactions, and a premium purchasing journey. Behind the scenes, we built a scalable e-commerce foundation capable of supporting Dorbi's ambitions to become a leading destination for supercar accessories and services in the region.
Working alongside the team at Narsun Studios, Grayhat contributed to Arkadium’s crossword gaming platform, maintaining and evolving a portfolio of over 30 crossword experiences built in React. These included globally recognized titles such as Daily Crossword, Daily Mini Crossword, Daily Commuter Crossword, Thomas Joseph Crossword, Stan Newman Daily Crossword, and Penny Dell Crosswords, serving a large audience of puzzle enthusiasts across multiple publishers and platforms.
Our Grayhatter took complete ownership of the crossword delivery pipeline, collaborating closely with Project Managers, Product Owners, DevOps engineers, frontend developers, and other stakeholders to ensure smooth releases and a consistent player experience. Beyond feature development, they navigated a complex legacy codebase built on Node.js 11 and older React architectures, taking responsibility for maintaining multiple variants of the same core application.
The work extended beyond day-to-day maintenance. The Grayhatter uncovered and resolved deeply rooted issues hidden within years of accumulated code, contributed to the migration toward more modern development practices, and helped refactor critical parts of the platform to improve stability, maintainability, and future development velocity.
Amazons Real is a real-estate marketplace platform built for the Australian property market, bringing multiple property-related verticals into one unified digital experience.
In collaboration with the client, we developed Amazons Real as a multi-surface platform covering real estate sales, rentals, commercial properties, hotels and resorts, auctions, news, and supporting property services. The core vision is to create a centralized destination where users can browse, search, compare, and engage with different property opportunities across the Australian market.
The system includes a public-facing website for property discovery, with flows for searching listings, applying filters, viewing property details, contacting agents, exploring auctions, reading news, and accessing account-based features through login and signup. Alongside this, the platform includes an agent portal where agents can sign up, authenticate, create listings, manage property states, and move through verification and application workflows.
Amazons Real also includes an administrative layer for managing users, properties, and agent applications, with a superadmin experience for broader platform oversight. The listing workflow supports states such as active, pending approval, expired or downgraded, inactive, pending payment, and unavailable or sold, helping the platform support real operational property lifecycles.
We designed and developed the landing page and documentation platform for Listen.dev, a modern supply-chain security startup helping engineering teams identify, understand, and resolve vulnerabilities throughout their software development lifecycle.
Built in Next.js with Framer Motion, the experience was crafted to feel familiar to modern developers, drawing inspiration from the best of developer tooling. We also connected a CMS to create content focused on actionable security education, technical blogs, and a frictionless onboarding experience.
Performance and developer experience remained central throughout development. We carefully optimized rendering strategies, reduced unnecessary client-side complexity, streamlined SVG and animation usage, and improved overall page performance while preserving the polished visual identity of the product.
Following the success of Sain Smart Notes, we identified that one of its most loved features—the AI prompt library—deserved a product of its own. AI Smart Chat was born as a standalone AI chatbot experience, allowing users to discover curated prompts, generate ideas, and engage with AI through a fast, focused interface.
Starting from the foundations of Sain, we rapidly transformed the architecture into a dedicated AI chat application, preserving core systems like authentication, AI integrations, subscriptions, and prompt management while redesigning the experience around conversation-first workflows. The application featured AI chat history, categorized prompt discovery, premium subscriptions, and a streamlined mobile-first experience.
Built in Flutter with Firebase powering authentication and cloud storage, the platform integrated modern AI APIs alongside subscription systems and app-store compliant purchase flows. Despite beginning with limited design resources, our team adapted quickly to craft a polished user experience and successfully brought the standalone AI product to market.
Some engagements are remembered for what was built. Others are remembered for the decisions made when a project’s technical direction, legal exposure, and professional boundaries no longer aligned.
This was one of those projects.
The client approached Grayhat to develop highly specialized ticket automation infrastructure involving proxies, anti-bot circumvention techniques, and interaction with systems protected by technologies such as PerimeterX. The initial scope presented itself as an advanced automation challenge, as was in the name itself. However, as the engagement progressed and our engineers learned more and more about the requirements post-agreement, it became evident that the proposed methods relied heavily on unstable tooling (which the client themselves provided and insisted on using) and increasingly questionable approaches (and close to impossible, without jumping into possibly illegal areas) that our team was not comfortable deploying.
We attempted to redirect the project toward more reliable and professionally maintainable technical solutions. Those recommendations were rejected. The relationship quickly deteriorated, and after a short period of engagement, the client publicly accused our team of incompetence, misrepresentation, and contractual breach, despite the fact that the project had only progressed through an initial paid engagement period for which engineering time and resources had already been committed.
It is important to understand the timeline.
The client’s public review described imminent legal action against Grayhat. No such legal action was ultimately brought. The advance payment in dispute represented compensation for engineering time already spent investigating, architecting, and evaluating the requested solutions—not a refundable deposit held without work being performed.
We have always believed that professionalism means saying a strong “no” when a request requires us to compromise on engineering standards, maintainability, or ethical boundaries.
This remains the only one-star review in Grayhat’s Upwork history. Every other project, before and after this engagement, has been built on long-term trust, technical excellence, and transparent communication with our partners. We discontinued our use of Upwork not soon after, because of other decisions, as the platform does not serve our interests.
In retrospect, this project became a defining lesson: the best engineers are not only measured by what they can build, but by the problems they refuse to solve the wrong way.
Important Update for further context: As of 2026 (the time of writing this), the client's company has been shut down due to this very same questionable and unethical behaviour, by the same company against which the client had requested us to do questionable activity on. It brings Grayhat no satisfaction, however, it does teach us a lesson, humbles us, and helps us align ourselves in the future to take on work which has better impact on peoples' lives.
Listen was built around a simple belief: security tools should feel like developer tools. Designed as a modern platform for software supply chain security, it integrates directly into development workflows to surface actionable security insights where teams already work—inside pull requests, repositories, and deployment pipelines.
Working alongside an international team of eBPF and security engineers from Italy and a business team in Canada, our Grayhatters lead frontend development and helped bring the initial platform to life in under 2.5 months. The MVP combined GitHub-native workflows with a Vercel-inspired user experience, transforming dependency and package security analysis into verdicts which developers could actually understand and act upon.
The platform synchronized directly with pull requests, analyzed dependency trees and package manifests, and surfaced detailed security breakdowns through an interface designed to reduce alert fatigue and improve remediation speed. Inspired by the best parts of GitHub and Vercel, the experience focused on clarity, discoverability, and seamless developer adoption.
Our work included building a polished Next.js frontend, crafting high-fidelity interactions with Framer Motion, and implementing the complete GitHub-centric onboarding, navigation flow and DX. The result was a security product that felt less like enterprise software and more like a tool developers genuinely wanted to use.
Heedy is a social and monitoring platform built to bridge the gap between generations, helping families stay continuously connected with elderly relatives through simple, reliable technology.
In a deep collaboration with entrepreneurs in Sweden, we developed Heedy into a production-ready system that was also pitched to local municipalities as a digital health and inclusion solution. The core vision is to reduce digital exclusion, isolation, and loneliness by making communication effortless across all ages—and by giving older or obsolete devices a second life within a meaningful use case.
The system runs across both iOS and Android, with a dual-experience design: a passive, always-on “elderly view” on tablets that requires no interaction, and an active family-side app for video calls, messaging, and shared updates. This structure enables continuous presence without demanding technical literacy from elderly users.
The product has been tested in the real world, with early users across mobile platforms and feedback from families validating its direction as a critical tool for care, connection, and aging-in-place support.
A cross-platform calling platform built for professionals whose time is their product. Designed for lawyers, consultants, healthcare providers, and other service businesses, the app automatically tracks billable call time and streamlines the path from conversation to invoice.
We led the project from product design through engineering, creating a complete Figma system and crafting an experience that closely mirrors the native Android calling workflow while remaining fully cross-platform across iOS, Android, and desktop. Built with Flutter and Material 3, the platform includes voice calling, messaging, contact synchronization, calendar scheduling, project dashboards, do-not-disturb controls, and the compliance requirements expected from a professional communications tool.
Under the hood, we engineered a scalable backend architecture in NestJS capable of supporting growth from an initial Australian market launch to global expansion. The result is a communication platform that feels familiar enough to replace a phone app, while adding the business workflows professionals need to monetize their time.
Sain (Smart AI Notes) was built to rethink personal productivity by combining note-taking, task management, and AI assistance into a single intelligent workspace. Designed for modern users, the platform transforms simple notes into a collaborative experience with AI-powered suggestions, content generation, and smart organization.
We led the complete mobile application development, crafting a polished cross-platform experience with rich note editing, to-do management, AI-powered writing assistance, and a curated prompts system. Users could format text, attach images, create checklists, manage tasks, and seamlessly import AI-generated ideas directly into their workflow.
Behind the scenes, we engineered a scalable cloud architecture using Firebase for authentication and data storage, integrated OpenAI for intelligent assistance, and implemented subscription-based monetization through Stripe, Apple App Store, Google Play Billing, and RevenueCat.
The result was a premium AI productivity application that brought together the emerging capabilities of generative AI with the everyday utility of notes and task management—creating an intelligent companion that helps users think, organize, and create more effectively.
Only Sports was a social platform built around the emerging NIL movement, allowing college athletes to connect directly with their fans, showcase their journeys, and generate income through their content. The platform combined social experiences, creator monetization, and athlete discovery into a new digital ecosystem for student sports.
Working as part of Antematter’s engineering team, our Grayhatter contributed to the development of the web platform within a highly distributed microservices architecture spanning multiple services and technologies. Navigating a complex legacy codebase, they took ownership of challenging frontend initiatives, untangled technical bottlenecks, and consistently helped the team move forward during critical development phases.
Beyond implementation, the engagement demanded deep system-level understanding, cross-service collaboration, and the ability to work effectively across a large engineering surface area. It became a defining experience in handling enterprise-scale architectures, complex integrations, and the realities of developing products built on distributed systems.
HandyApp is a contractor focused workforce platform built to help service professionals manage onboarding, job activity, documents, communication, and scheduling through one connected mobile experience, and for customers to look for and hire such professionals.
In collaboration with the HandyApp team, we worked on the contractor side mobile application, supporting flows that help users create their account, register company details, select offered services, manage required documents, and stay connected with active work opportunities. The core vision is to make contractor onboarding and daily work management simpler, more organized, and easier to operate at scale.
Following the success of PlayroomKit for Unity, we were tasked with bringing the platform to Apple's ecosystem through a native Swift integration. The goal was simple: enable developers to build multiplayer experiences using native Apple technologies while paving the way for emerging platforms like VisionOS.
We engineered a lightweight native bridge that wrapped the existing JavaScript SDK inside a WebView layer, exposing PlayroomKit functionality through a clean Objective-C and Swift interface. We validated across iOS and macOS.
While the project was ultimately shelved, it demonstrated a viable path for bringing real-time multiplayer capabilities to the broader Apple ecosystem—and laid the groundwork for what PlayroomKit could become beyond the browser.
Following the breakout success of Booom.ai and the rapid rise of AI-powered games, we helped Playroom build AIGames.gg (later rebranded as dpad.ai) as a dedicated discovery platform for the growing ecosystem of AI-native games—one of the first AI game directories, before they became commonplace.
At a time when AI gaming was still fragmented and difficult to discover, the platform gave developers a place to showcase projects and players a way to explore the next generation of interactive entertainment, and vote on their favorite ones to bring them to the top.
Built with Next.js static site generation for exceptional performance, the platform combined a clean neobrutalist design system with fluid Framer Motion animations and transitions, creating an experience that felt more like a modern product than a traditional content site.
Behind the scenes, we developed a complete publishing and monetization system, including a completely custom admin panel for content management, paid submissions, editorial workflows like Editor's Picks, and post creation.
Playroom began as an ambitious vision to bring social gaming back to the living room. Players would scan a QR code with their phones, instantly join a shared room, and use their devices as controllers while the game unfolded on a TV or shared screen.
Working alongside the founding team of a FAANG designer and engineer duo, we helped build both the platform and the experiences that brought it to life. From reaction-based party games and competitive tap challenges to airplane games controlled by the phone as a controller, and other multiplayer classics like Rock Paper Scissors, we explored what phone-powered social gaming could become.
Beyond the games themselves, our team member contributed to the core multiplayer technology powering room creation, synchronization, and real-time phone-to-screen interaction. The lessons learned from building the platform ultimately evolved into PlayroomKit—the standalone multiplayer SDK that became the company's flagship product and now powers hundreds of multiplayer experiences.
ChatBuddy is a multi-model AI assistant built to unify the fragmented LLM ecosystem into a single, fast, ad-free interface. It brings together leading models including ChatGPT, Claude, Gemini, Grok, DeepSeek, Mistral, Venice AI, Llama, and more, allowing users to switch seamlessly between providers inside one consistent experience.
We helped establish the initial architecture and prototype of the app, focusing on multi-model routing, unified conversation handling, and a clean, native-first UX designed for Apple platforms. The goal was to abstract away model complexity and let users interact with AI as a single coherent system rather than a collection of disconnected tools.
Before AI gaming became a category, we made Booom.ai, turning generative AI into the game itself. Players could enter any topic and instantly generate a multiplayer trivia experience, creating effectively unlimited gameplay powered by GPT 2. The concept was so novel at the time that it stood among the earliest examples of AI being used as a core game mechanic rather than a supporting feature. (featured #26 on ProductHunt, on July 18, 2023)
We helped shape the experience through a distinctive neobrutalist design language, polished multiplayer flows, and rapid iteration that helped the game reach more than 50,000 players across the globe within its first months.
The platform later expanded through a partnership with MiniParty (formerly Bunch), bringing AI-powered social gameplay to a wider multiplayer audience.
As the product matured, it evolved beyond generated quizzes into a full creator platform, allowing users to build and share their own custom trivia experiences through an AI-powered canvas.
What began as an experiment in GPT-powered entertainment became an early proof that AI could create entirely new forms of interactive play years before the broader industry caught up.
TrainEngine.ai was a first-of-its-kind platform for creating, training, and applying custom image models built around Dreambooth-style workflows. It enables artists and creators to fine-tune their own models on demand and generate large volumes of AI-driven visual assets from them.
The platform is designed with a GitHub-like mental model: GitHub authentication, one-to-one user mapping, and a familiar repository-style structure for managing models and outputs. Beyond single-model generation, it introduces model chaining—allowing multiple trained models to be composed together to produce richer, more expressive results.
We built features for end-to-end creative control, including a UI for model fine-tuning, generative asset creation, and seamless experimentation across models. The result is a system that blends infrastructure-grade AI tooling with a creator-first interface, aimed at expanding what’s possible in digital art and design.
Following the success of Booom, Playroom expanded its AI-powered gaming experiments with Fax, a social game where players had to separate genuine facts from convincing AI-generated lies. The project explored how emerging generative AI technologies could create entirely new forms of interactive entertainment.
The Grayhatter behind Fax owned the complete AI experience end-to-end, developing the AI backend, integrating one of the earliest ElevenLabs-powered voice generation pipelines in gaming, and building the frontend systems that brought the AI personalities and interactions to life.
By combining AI-generated voices, content generation, and real-time multiplayer experiences, Fax demonstrated how cutting-edge AI could move beyond productivity tools and become a foundation for playful, engaging consumer experiences.
The MegaLecture Mobile Application is an EdTech platform developed for MegaLecture, an educator-led client focused on helping college students prepare more effectively for board examinations through structured digital learning.
In close collaboration with the MegaLecture team, we developed the mobile application using Flutter and Firebase, with support for both iOS and Android. The platform was designed around a clear academic goal: to help students organize, learn, and practise for their college papers across Cambridge International O & A Levels, IGCSE, Pakistan Matric, and FSC.
The application brings together handwritten notes, video lectures, complete examination past papers, MCQ papers, and theory papers from well-reputed tutors and instructors for each supported board. Students can filter papers and resources by board, year, session, topic, and paper type, making it easier to focus on the exact material relevant to their syllabus and exam preparation.
A core part of the product is its interactive practice experience. Students can attempt MCQ questionnaires with real-time answer feedback or switch into a paper simulation mode with no feedback, creating a more realistic exam environment. For theory papers, students can submit image-based answers for tutor review. Each question is supported with relevant notes, video lectures, and resource links, allowing students to learn directly from the material connected to their practice.
JPN International is a family-owned Japanese automotive and marine business based in Nagoya, connecting customers worldwide with direct access to Japanese vehicle auctions, marine products, and export services through a transparent, technology-first experience. The company aimed to modernize how international customers discover its services.
We designed and engineered a prototype for a complete digital platform that balanced modern minimalism with Japanese design sensibilities. Built with React and Django REST Framework, the system included a custom vehicle marketplace, auction inquiry workflows, guest and authenticated shopping carts, Stripe-powered payments, multilingual support, and a complete inventory management backend.
The experience was crafted to serve both local and international audiences, featuring seamless search and discovery across vehicle listings, marine inventory showcases, secure user accounts, and a streamlined checkout flow. Behind the scenes, a robust Django administration system provided custom management tools to handle inventory, customer interactions, and business operations.
Spoil:Me is a cross-platform social media application developed for iOS and Android, designed around photo sharing, video sharing, stories, and profile-based user interaction.
The platform was built around the familiar experience of modern social networks, allowing users to create profiles, publish posts, upload photos and videos, share stories, and engage with content from other users. A key part of Spoil:Me is its in-app coin system. Users can purchase coins inside the application and spend them on posts created by other users as a form of appreciation and support.
We helped develop the React Native frontend, and its integration with the backend, for both Android and iOS users.
NSIST (National Superior Institute of Science & Technology) was a new higher education initiative in Islamabad, backed by the Superior Group — one of Pakistan’s largest educational networks behind institutions such as Superior University. As the institute prepared to launch its undergraduate programs, it required a digital experience capable of communicating its vision, programs, and unique academic positioning to prospective students.
Grayhat designed and developed a modern, accessible, and conversion-focused website that translated complex institutional messaging into a simple student journey. Built with Next.js and deployed on Vercel, the platform highlighted NSIST’s academic offerings, industry-focused learning philosophy, and admissions process through a fast, intuitive web experience.
The result was a clean digital identity for a growing educational institution — combining strong visual storytelling with practical information architecture to help students explore, understand, and take the next step in their academic journey.
Ticka is a TV-first financial market display built for Apple TV, designed to bring crypto, stocks, indices, and forex into a single, elegant, always-on interface. It transforms living rooms, offices, trading desks, and public spaces into real-time market environments with a clean, readable design.
We built Ticka in Swift for Apple TV, using the best practices of Apple TV—large-format pricing, minimalist dark themes, and smooth Apple TV-native navigation using the remote. It’s optimized for environments where information needs to be visible instantly, from across the room.
The app delivers low-latency market updates with support for major assets including Bitcoin, Ethereum, Solana, Cardano, Dogecoin, global indices, and equities across NASDAQ, NYSE, and ASX. Users can create custom watch pages, view multi-timeframe charts, and monitor live tickers and market context at a glance.
We also helped build out the premium tier offering expanded pages, low-latency quotes, and advanced monitoring capabilities for more demanding setups.
Kontrol was a rider management and supply chain coordination platform developed for Salon Dziners, one of Pakistan's largest cosmetics distributors. It enabled real-time task distribution, delivery tracking, payment logging, and communication between field riders and administrators. Built as a native Android application in Kotlin with a web-based admin dashboard, the platform digitized delivery workflows, reduced paperwork, and improved operational visibility across Salon Dziners' distribution network throughout Pakistan's Twin Cities.