Hi, I'm Tushar.

Open to Work

Applied AI Engineer building RAG pipelines, intelligent AI agents, AI products, and high-performance backend systems.

About

Applied AI Engineer focused on building production-ready RAG pipelines, LLM-powered applications, intelligent model routing, and reliable backend systems. Experienced with Python, TypeScript, API development, AI evaluation, and backend reliability.

Open-source contributor to NanoCollective, Cia ren Labs, and sklearn-genetic-opt, contributing to AI/ML libraries and developer tooling.

Work Experience

Avoir

Aug 2026 – Sep 2026 · Remote

  • Fixed IDOR vulnerabilities by enforcing authenticated identity from verified Cognito JWTs across protected routes.
  • Prevented concurrent credit-billing bypasses with atomic, fail-closed quota enforcement.
  • Built a three-tier trend-data pipeline with fallbacks, circuit breaking, and DynamoDB caching.
  • Implemented Stripe checkout, webhooks, subscription transitions, and an audit ledger.
  • Improved delivery workflows with CI/CD checks for linting, type-checking, tests, and secret scanning.

Open Source Contributions

  • Built and integrated core tuning/evaluation utilities to improve reproducibility of LLM experimentation workflows.
  • Improved developer experience with cleaner APIs, better error handling, and test coverage for critical paths.
  • Refactored internal modules for modularity, making them easier to extend and compose.
  • Implemented and extended backend functionality to support AI/agent workflows with cleaner, typed interfaces.
  • Hardened validation and error flows to make services more predictable and production-friendly.
  • Contributed to developer tooling and integration points to streamline local/dev workflows.

Skills

PythonTypeScriptGoNode.jsFastAPIReactNext.jsPostgreSQLDynamoDBRedisDockerKubernetesAWSCI/CDGitGitHub ActionsREST APIsRAGHybrid SearchEmbeddingsRerankingLLM EvaluationModel RoutingAI AgentspytestLinux

Projects

llm-gateway

A multi-provider LLM gateway — one OpenAI-compatible API in front of Groq, Gemini and OpenRouter, with auto-routing, caching, failover and a Next.js dashboard to monitor cost and usage.

luman

A retrieval-augmented generation engine over your own documents. Point it at a folder, ask questions, get answers that cite the exact passage they came from — and an honest "the documents do not say" when the evidence is missing.