Agents, RAG, and the full stack around them. My open-source and production work ships to 70+ organisations, and I own the whole loop: problem → architecture → CI/CD → observability in production.

🌐 siddharth-dev.tech  ·  📍 Remote — India (UTC +5:30)  ·  💬 Open to senior remote roles

Avni Skills SDK — open-source agentic SDK · core engineer · adopted org-wide

• A chat-driven generator wrapped in a Claude-Agent loop where every turn is a git commit and a validator runs after it.
• Domain rules enforced as MCP tool-gates the agent must call — plus a hook that auto-reverts anything it does off-script.
• Hard $5/session cost cap and a thrash-detector; 463 unit tests + a real-LLM eval harness that scores agent quality release-over-release.

Avni AI — production RAG + LangGraph · core engineer

• Turns an NGO's programme spec into a deployable, field-ready mobile app — forms, validation rules, and logic generated by RAG + LangGraph, then human-reviewed.
• Cuts implementation from 2–3 weeks to ~1 day; ships to 70+ orgs and 11M+ beneficiaries. Recognised as a Digital Public Good.

Avni Support Ops — agentic support workspace · sole author · private

• Read-only / draft-only by construction: the AI drafts, a human decides — there is no write-DB connection anywhere in the workspace.
• −60% ticket resolution time and +40% team efficiency, measured on real production tickets.

• QuickPe — real-time fintech wallet · ACID-compliant transactions, RBAC, JWT auth, audit trails · scaled to 2,190+ concurrent users at ~391 ms, −70% DB load.
• MeTube — video streaming platform · −92% API calls via caching & debouncing · real-time chat at <35 ms latency.
• OpenCam — offline-first browser camera suite · MediaDevices API + IndexedDB, multi-device & cross-browser.

Languages

Frontend

Backend

AI & Agents

Data

Cloud & Ops

🌐 siddharth-dev.tech  ·  💼 LinkedIn  ·  𝕏 @siddharthraj29  ·  ✉️ [email protected]

🔥 I build AI that ships — and stays shipped: effective, scalable, and measured.

_{Also on GitHub: @siddharthr29}