Our founder worked as a fullstack developer for ProdigyBuild, focusing on creating intuitive and responsive user interfaces with backend services integrations. ProdigyBuild is an AI-powered tool for software teams that streamlines task management, estimation, and documentation—reducing time-to-market and developer fatigue through its intelligent project management capabilities.
Product teams struggled with planning accuracy, documentation backlogs, and fragmented dev workflows.
Our founder contributed to building the marketing website and refining how the platform communicates its core value to prospective users and developers.
Fullstack development
Intuitive user interface design
Responsive web components
Backend service integration
User authentication flows
Task management systems
Project estimation tools
Documentation automation
Developer workflow optimization
Ensuring responsive and intuitive user interfaces that could handle complex project management data without performance degradation.
Maintaining consistent design language and component behavior across multiple sections of the application.
Creating interfaces that worked seamlessly across desktop, tablet, and mobile devices while preserving full functionality.
Developing intuitive visualizations for complex project metrics and task relationships that remained clear and useful.
Created a comprehensive component library with standardized behavior and styling to ensure consistency across the application.
Implemented code-splitting, lazy loading, and optimized rendering techniques to ensure smooth performance even with complex data visualization.
Developed a flexible grid system and adaptive components that intelligently reorganized based on screen size and device capabilities.
Conducted regular usability testing with software development teams to refine interfaces based on real-world usage patterns.
Built accessibility considerations into the core development process, ensuring the application was usable by team members with diverse needs.
ProdigyBuild is an AI-powered project management SaaS where an assistant expands brief issues into detailed tasks with implementation notes, test steps and time estimates. The product runs as a serverless Node.js backend on AWS with Cognito authentication, PostgreSQL and Redis, and an OpenAI-backed enrichment layer, fronted by a React and Ant Design web app. Serverless fit a bursty, per-tenant workload and let a small team avoid managing servers.
A single-page React app for boards, estimates and documentation, plus a marketing and onboarding site.
Static assets ship from a CDN while an API gateway authenticates every call with Cognito-issued tokens.
Business logic is split into serverless functions grouped by domain, invoked synchronously for the UI and asynchronously for AI jobs.
Relational data for projects and tasks, a cache for hot board state, and object storage for generated documents.
An OpenAI-backed layer rewrites issues, proposes implementation and test steps and estimates effort; external tools plug in via APIs.
Infrastructure defined as code and deployed through CI with static analysis gates.
A user signs in through Cognito and loads a project board; the React app fetches issues via the API gateway.
The user creates a one-line issue and asks the AI assistant to expand it.
A function assembles project context, relevant code and sprint history into a prompt and sends it to the OpenAI API asynchronously.
The response is parsed into description, implementation notes, test steps and an estimate, then stored in PostgreSQL.
Redis invalidates the board cache and the UI refreshes with the enriched task and updated gantt projection.
Repository events from the connected VCS arrive via webhooks and update task status and release progress.
Usage is bursty and tenant-driven, with AI calls that can take seconds; pay-per-invocation functions scale to zero between sprints and isolate slow AI work from UI requests.
Calling the OpenAI API inline would block the UI and hit rate limits; queued jobs with Redis-backed status keep boards responsive and allow retries.
Organisations need SSO-ready sign-in and role scoping without the team owning password storage; Cognito supplied this with JWTs the gateway can verify at the edge.
Boards and gantt views are data-heavy, so a shared Ant Design-based component library with lazy-loaded routes kept rendering fast and visuals consistent.
“The website development work has significantly improved our user engagement and conversion rates, effectively communicating our product's value to potential customers.”
We ship MVPs, AI agents, and production systems in 8 weeks. Bring the rough spec — we'll scope the build inside a week with a fixed timeline and quote.
