Full Stack Developer from India who likes shipping things that actually run in production — UI, API, database, deploy, all of it. Spent most of my time in the MERN stack and Next.js, and lately I've been going deep into Python and AI: LLM apps, tool calling, retrieval, and agents that do real work instead of just chatting.
I care about the boring parts too — latency, cost per call, retries, and what happens when the model gets it wrong.
agent:
name: deepanshu
role: Full Stack Developer
location: India 🇮🇳
status: building
system_prompt: >
Turn ideas into working products.
Prefer shipping over planning.
Automate anything done twice.
tools:
- typescript # react, next.js, node
- python # fastapi, scripting, ml
- react_native # mobile
- llm_apis # prompting, tool calling, rag
learning:
- agent architectures & orchestration
- machine learning fundamentals
- evals: measuring what "good" means
open_to: [open source, ai projects, collabs]Always up for talking about AI agents, product ideas, or anything worth building.

