HRITIKRAJAI / PRODUCT
Full-stack AI engineer building complete products across React interfaces, FastAPI services, LLM orchestration, secure multi-tenant data, automated testing, and cloud delivery.
01Selected systems
All projectsFull-stack AI engineer · 2025–26
SupplierVerse.ai
An AI procurement platform that turns fragmented public tenders into grounded, reviewable bid work across discovery, matching, collaboration, and response drafting. Worked across the Next.js product, 85 FastAPI endpoints, 297 database migrations, ingestion jobs, matching, collaboration, billing, and shared identity boundaries. Built the Azure OpenAI and LangGraph Bid Writer flow around typed state and reviewable output.
Full-stack developer / AI engineer · 2025–26
GoDiverse
A multi-persona B2B operating system spanning buyer discovery, supplier operations, verification, teams, controlled public data, and shared identity. Delivered discovery, supplier profiles, verifier sessions, team administration, safe public projections, and a shared app-access migration with compatibility fallback, backfills, and orphan repair.
Product engineer · 2026
UplyFox
A browser extension and web control plane designed to make application assistance useful without making submission unsafe. Built grounded profiles, deterministic answers, form inspection, verified writes, attachment handling, tracking, and explicit submission approval.
Product engineer · 2026
AI Open Studio
A collaborative learning and research workspace combining a Mistral prompt engine, structured document AI, shared Yjs state, and native peer media. Built 30 prompts across 10 categories, constrained BlockNote operations, research rooms, a custom Yjs transport with 50 ms batching, reactions, drawings, and native WebRTC signalling.
Product engineer · 2026
AffairsAI
A bilingual current-affairs product that turns resilient public-feed ingestion into daily briefs, validated quizzes, research digests, and offline study tools. Built collection across 119 registered source variants, cleaning, deduplication, scheduled daily briefs, validated quizzes, bilingual generation, SEO, and an offline-first Expo client.
Product engineer · 2026
Coders Cookies
An open developer learning platform with 504 DSA problems, structured roadmaps, multi-language code execution, safe public profiles, and a published theme suite. Built DSA navigation, 150 editorials, 630 checkable topics, a Monaco-powered runner, resume tooling, public profiles, and a 12-theme VS Code Marketplace extension.
02Engineering range
From interface
to operated system.
I own the path from product interaction and API design through AI orchestration, secure data boundaries, testing, and deployment. That range lets me connect model behavior to evidence, automation to human approval, and ambitious interfaces to infrastructure that can fail honestly.
AI systems
Graphs, tool-calling, structured outputs, retrieval and evidence controls
02Product security
RLS, tenant scope, identity migration, approval-bound automation
03Product engineering
React, Next.js, FastAPI, API design, browser extensions and rich editors
04Delivery
Docker, Azure, AWS, CI/CD, Playwright and operational fallbacks
03Independent experiments
GitHub Wrap / visual engineering
Data made tangible.
A contribution-history experiment that turns GitHub activity into heatmaps, fleets, skylines, ocean currents, and a Bengaluru traffic visualization.
04Field notes
All writingEvidence before generation
Designing a bid-writing graph that cannot silently turn model confidence into unsupported claims.
02Approval bound to state
Why a click on “approve” should authorize one exact form—not a future page that merely looks similar.
03Realtime, with honest limits
A custom Yjs transport and native WebRTC room are useful engineering—not permission to claim infinite scale.