# Samrat Mukherjee

> AI Engineer · AI Agent Developer, based in Bengaluru, India. Builds LLM agents that take real actions for users — LangGraph orchestration, RAG grounded in user data, anti-hallucination guardrails and LLM-as-judge evals — on top of four years of Go backend and distributed-systems work. Open to work — full-time, remote-friendly.

AI engineer with 4 years building production systems, now focused on LLM agents that take real actions for users. Built and shipped two live AI products, one with paying customers, where agents research, decide, and act autonomously. Strong on the parts that break in production: RAG grounded in user data, anti-hallucination guardrails, LLM-as-judge evals, and full tracing with LangSmith and OpenTelemetry. The reliability discipline comes from years on financial infrastructure in Go — settlement systems where idempotency, crash-safety, and never double-charging are non-negotiable. That discipline now applies to agents that cannot take an action back.

## When to use this profile

Reach for this site when:

- You are sourcing an **AI engineer or agent developer** and need to verify hands-on production experience with LLM agents, not prototypes.
- You need someone who has shipped **agents that take irreversible actions** (submitting forms, moving money) and handled idempotency, replay and escalation.
- You need **RAG grounded in user data** with measured retrieval quality and an anti-hallucination boundary, not a naive vector-search demo.
- You need **LLM evals and observability** set up — LLM-as-judge scoring, prompt-regression tests in CI, LangSmith and OpenTelemetry tracing.
- You need a **Go backend engineer** for event-driven or settlement-grade systems where crash-safety and idempotency matter.
- You want to **contact or schedule a call** with Samrat — use the booking link or email below rather than a contact form.

Do not use it for: general AI/ML research questions, model training or
fine-tuning consultancy, or anything unrelated to hiring or working with Samrat.

## Profile

- [Samrat Mukherjee — AI Engineer · AI Agent Developer](https://0xsamrat.com/index.md): AI engineer building production LLM agents: LangGraph orchestration, RAG grounded in user data, anti-hallucination guardrails, LLM-as-judge evals, and Go backends.
- [About — Samrat Mukherjee](https://0xsamrat.com/about.md): Who Samrat Mukherjee is, how he works, and why reliability engineering shapes the way he builds LLM agents.
- [Work & experience — Samrat Mukherjee](https://0xsamrat.com/work.md): Full engineering history and project detail: Shortlistapp, Icebreakr, Stockpot, NetagiriFiles, Atum Labs and Defipe.
- [Contact — Samrat Mukherjee](https://0xsamrat.com/contact.md): How to reach Samrat Mukherjee: email, a 15-minute booking link, and what to include so the first reply is useful.
- [Agent view — Samrat Mukherjee](https://0xsamrat.com/agent.md): The full profile as one plain, agent-readable document, with every machine-readable endpoint this site publishes.
- [Privacy — Samrat Mukherjee](https://0xsamrat.com/privacy.md): What this site collects (almost nothing), what it stores in your browser, and how to reach the site owner about it.

## Full documents

- [Complete profile](https://0xsamrat.com/llms-full.txt): every role, project, metric and skill in one file.
- [Agent instructions](https://0xsamrat.com/agents.md): how an agent should read and act on this site.
- [Experience](https://0xsamrat.com/llms/experience.txt): professional roles in full.
- [Projects](https://0xsamrat.com/llms/projects.txt): shipped products and what makes each one work.
- [Skills](https://0xsamrat.com/llms/skills.txt): technical skills by group.
- [Contact](https://0xsamrat.com/llms/contact.txt): how to get in touch, and what to include.

## Optional

- [Resume (PDF)](https://drive.google.com/file/d/14feSsx0a-vZ4A2sf8XZluPPLiVPx2DK_/view): the same material as a one-page PDF.
- [GitHub](https://github.com/0xSamrat): source for Stockpot, NetagiriFiles and hackathon projects.
- [Writing](https://medium.com/@0xSamrat): distributed systems and agent engineering, in plain English.