Work

LoomKit

Featured
2026

Tiny, swappable LLM orchestration core — the seam under your agent harness, without LangChain-scale bloat. A Protocol, a few dataclasses, one optional provider.

  • Provider Protocol — swap backends without inheritance
  • Small Message · Response · Usage surface
  • Gemini out of the box; core stays zero-deps
  • Ready to layer streaming, tools, and agents

FeedFr

Live
2026

User feedback that ships. Collect ideas, prioritize with votes, show what's planned, and announce what landed — without enterprise pricing or clutter.

  • Public feedback boards with voting
  • Roadmaps and changelogs that close the loop
  • Embeddable widget for in-product capture
  • Free forever tier — paid when you outgrow it
feedfr.comNext.js · PostgreSQL

Path

  • 2025 — Present

    Tudip Technologies

    · AI Engineer

    Intelligent pipelines and LLM integrations for enterprise applications.

  • 2020 — 2022

    IBMz

    · Z Ambassador

    Enterprise computing advocacy and technical workshops.

Tools

Languages
JavaScript, TypeScript, Python, Go
AI
LLMs, LangChain, LangGraph, RAG, TensorFlow, PyTorch
Infrastructure
PostgreSQL, MongoDB, Redis, Docker, Kubernetes, AWS
Frameworks
FastAPI, Next.js, Node.js, Flask, Pydantic, SQLAlchemy

Contact

Open to thoughtful collaborations and interesting problems.

If you're building something that needs AI depth, backend architecture, or a product-minded engineer — write.

jadon.devyanshu@gmail.com

© 2026 Devyanshu Jadon