⌘resume

i'm an ai engineer who builds llm-powered products — rag pipelines, multi-provider llm tooling, and the python apis behind them. i like turning messy problems into small, fast tools people actually want to use.

• emailt.prayag.eng@gmail.com
• github@prayagtushar
• linkedin@prayagtushar
• websiteprayagtushar.xyz

• metquay inc · software engineer

  june 2025 - present · trivandrum, india
  • part of the core team rebuilding the main metquay product on a modern stack (react, nestjs, and turborepo), moving it off a legacy codebase feature by feature.
  • built a sticker configuration panel with fabric.js, letting users design and edit stickers directly on a web canvas.
  • worked on certificate generation and pdf handling, including its integration with the sticker tool.
  • currently building an offline mode so the full application can run as a pwa without an internet connection.
• zerone consulting · software developer

  jan 2025 - may 2025 · kochi, india
  • developed a full tuition management application from the ground up, implementing a complex system for hierarchical user access.
  • built a secure authentication and role-based access control system using jwt to ensure strict data privacy for students, teachers, and admins.
  • designed and tested rest apis to facilitate seamless communication between the frontend and backend, ensuring reliable database storage.
• collab junction · application developer

  june 2024 - aug 2024 · vellore, india
  • improved platform stability by implementing enhanced error messaging and logging systems, successfully reducing application crashes by 30%.
  • optimized website performance and google search ranking (seo), leading to a 20% increase in daily visitor traffic.

• multi-llm client — unified async client for openai, anthropic & gemini05.2026code

  a provider-agnostic async python client that normalizes messages, streaming, token usage, and errors across openai, anthropic, and gemini behind one interface. ships as a library, cli, interactive repl, and fastapi service — with tenacity retries, pydantic v2 models, mypy-strict typing, a concurrent compare() across providers, and 40 tests.

  pythonasyncioopenaianthropicgeminifastapipydantictenacity
• readora — rag chat interface for pdfs05.2025livecode

  built a citation-style rag chat app where users upload pdfs and ask grounded questions. pipelines gemini-embedding-001 into pinecone (namespace per file), streams answers via gemini-2.5-flash, and stores files on vercel blob with neon postgres + drizzle orm.

  next.js 15gemini 2.5 flashpineconevercel blobclerkdrizzle ormneon postgresbun
• ask video.ai — saas rag bot for youtube12.2024livecode

  developed a web tool for video querying using langchain, gemini api, and pinecone. handled transcript chunking and embedding storage, resolving data overlap issues for real-time responses. managed codebase with turborepo and bun.

  next.jslangchaingemini apipineconeturborepobunprismapostgresql
• lumosai — chatbot powered by gemini08.2024livecode

  created a chatbot with gemini api for conversations, tested over 100 sessions to refine response accuracy. added clerk login and mongodb for chat storage. used tailwindcss for a mobile-friendly ui.

  next.jsgemini apiclerkmongodbtailwindcss