Agent orchestration
Stateful multi-step workflows in LangGraph that plan, call tools, recover on step failure, and escalate to a human when they cannot proceed safely.
Open to work — full-time, remote-friendly
I build LLM agents
that take real actions
for users.
I'm Samrat — an AI engineer in Bengaluru. Two of my agent products are live, one with paying customers. I work on the parts that break in production: retrieval grounded in real user data, guardrails that stop a model inventing facts, and evals that catch a regression before it ships.
- FocusAgent orchestration · RAG · evals & observability
- StackPython · LangGraph · Pinecone · FastAPI · Go
- Before this4 years of Go settlement & trading infrastructure
Samrat MukherjeeBLR · 2026
Agents in production
88%answer coverage
Outcomes
88%autonomous answer coverageA RAG rebuild of the answer layer at Shortlistapp raised how often the agent answers an employer's open-ended question on its own.
28%lower LLM costSame rebuild — retrieving facts instead of re-prompting a large model per question.
749tests on StockpotIncluding an adversarial suite against the agent's tool boundary.
~15×fewer Redis commands2.6M → 173K per month, by replacing a polling scheduler with a Lua timer wheel.
How I build agents
The interesting part is never the prompt — it's the boundary around the model.
An agent is only as useful as the worst thing it does unsupervised. Four things I get right before anything ships.
Grounded retrieval
RAG over the user's own data so answers come from real background, not from a model improvising. Retrieval quality measured, not assumed.
Evals & observability
LLM-as-judge scoring, prompt-regression testing in CI, and every model call, tool invocation and decision traced end to end.
Reliability for irreversible actions
Idempotency keys, crash-safe replay and hard anti-hallucination boundaries — because an agent that submits a job application cannot take it back.
Selected work
Shipped, and running in production.
012026Live · paying customers
Shortlistapp
An agent that applies to jobs end to end.
It reads the posting, fills the application, answers the employer's open-ended questions from the user's real background, and submits on real ATS portals.
88%autonomous answer coverage
022026Live on Chrome Web Store
Icebreakr
AI outreach that researches a person before it writes.
Drafts personalized connection requests, DMs and cold emails using LLMs, behind a provider architecture with automatic fallback chains.
4-axisLLM-as-judge eval harness
032026Agentic Commerce Hackathon
Stockpot
Agents that trade surplus food between restaurants.
Settling on real payment rails, where money moves only after delivery is verified.
749tests, incl. adversarial suite
042026500+ monthly visitors
NetagiriFiles
Civic transparency for ~800 Lok Sabha MPs.
Surfaces self-declared affidavit data with interactive D3.js charts, an India choropleth and per-MP profile pages.
<10 minfull dataset refresh
Experience
Four years, four systems that had to stay up.
Jan 2026 PresentCurrent
AI Engineer & Founder at Shortlistapp.co
Live AI agent product with paying customers. The agent applies to jobs end to end — it reads the posting, fills the application, answers the employer's open-ended questions from the user's real background, and submits.
- Built an autonomous job-application agent in Python (FastAPI, LangGraph) that plans and executes a full application as a stateful multi-step workflow, resolving each field through tool calls, recovering cleanly on step failure, and escalating to the user when it cannot proceed safely.
- Built a RAG pipeline over each user's resume and history (embeddings + Pinecone) so open-ended employer questions are answered from the user's actual background instead of generated from scratch. This raised autonomous AI answer coverage 88% and cut LLM cost 28% by retrieving facts rather than re-prompting a large model per question.
- Enforced a hard anti-hallucination boundary: a FactGuard layer blocks the model on any field with no retrieved fact, so the agent answers from real data or escalates. It never fabricates on a submission the user cannot take back.
Aug 2025 Nov 2025
Backend Engineer at Atum Labs (Atum Core)
Intent-driven, multi-chain stablecoin clearing & settlement network letting payment providers move value across chains through one API.
- Built core Go services for an intent-driven, auction-based settlement platform — payment gateway, solver gateway and on-chain verification — within a ~10-microservice system behind one REST API.
- Built an off-chain quote-auction pipeline on Apache Kafka using per-source topics plus ephemeral per-request correlation topics with TTL-based cleanup, decoupling services into an event-driven flow.
- Implemented a gRPC service (7 RPCs over Protocol Buffers) with a resumable server-streaming feed using monotonic sequence IDs, heartbeat liveness checks and replay-on-reconnect for reliable delivery across disconnects.
Sep 2023 May 2025
Founder & Lead Backend Engineer at Defipe.io
A derivatives trading protocol with a Go off-chain execution engine — order matching, position lifecycle, margin accounting and liquidation.
- Architected a Go off-chain execution engine for a derivatives trading protocol from scratch, covering order matching, position lifecycle, margin accounting and liquidation logic.
- Built a concurrent order-processing pipeline using goroutines for parallel matching and mutex-protected account state under simultaneous leverage constraints.
- Implemented the liquidation engine and funding-rate mechanism; validated across 10K+ simulated leveraged trades with zero false liquidations in testing.
Feb 2022 Nov 2022
Blockchain Developer at TechExactly
NFT-marketplace smart contracts and the React frontend that drove them.
- Engineered NFT-marketplace smart contracts with state-machine logic for ownership, settlement and transfer; built the React frontend with wallet connection and on-chain interactions.
- Implemented gasless approvals (ERC-2612 permit), removing a separate approval transaction to cut onboarding friction and user gas costs.
- Wrote test suites covering authorization, ownership transfer and failure paths to 95%+ functional coverage.
For agents
This site is machine-readable, on purpose.
I build agents that read the web, so this page answers Accept: text/markdown, publishes its own discovery catalog, and exposes in-page tools over WebMCP. No JavaScript required to read any of it.
$ curl -H 'Accept: text/markdown' \
0xsamrat.com
# Samrat Mukherjee — AI Engineer
> I build LLM agents that take real actions...
## Numbers
- **88%** — autonomous answer coverage.../llms.txtindex, with when-to-use guidance/llms-full.txtthe whole profile in one fetch/agents.mdhow to read, cite and act on this site/index.mdevery page has a Markdown twin/.well-known/ard.jsonAgentic Resource Discovery catalog/.well-known/agent-card.jsonA2A agent card
Open the agent viewOr append
?mode=agent to any URL on 0xsamrat.com.The journey
From 2022 to now.
Chronological — wins, founds, ships, and one shutdown that taught more than the wins did.
4years building
2×hackathon wins
$1kventure grant
2live AI products
2022
JoinedFeb 2022
Started at TechExactly
First gig in tech. Smart contracts, a React frontend, and the habit of writing tests for the failure paths.
2023
WinnerFeb 2023
Won ETH for ALL hackathon
First hackathon win — with the project that would become Defipe.
devfolio.coFoundedSep 2023
Founded Defipe
A derivatives trading protocol in Go. Off-chain execution engine, on-chain settlement.
WinnerSep 2023
Won ETHGlobal — Connext Pool Prize
Ape DAO 2.0: cross-chain governance with multichain proposal and voting execution.
ethglobal.com2024
Grant2024
Soonami Venturethon, Cohort 5
A $1,000 grant for Defipe, plus three months inside Soonami's network of founders and builders.
foundance.orgMilestone2024
Defipe → 500+ monthly active users
Peak usage, and a 1,000+ member Discord around it.
2025
Wound downMay 2025
Shut down Defipe
End of a two-year run. Took the lessons on idempotency and crash-safety with me.
JoinedAug 2025
Backend engineer at Atum Labs
Go services for stablecoin clearing and settlement — Kafka auctions, gRPC streaming, event-sourced Postgres.
2026
FoundedJan 2026
Started Shortlistapp.co
An autonomous job-application agent. First paying customers, and the first agent I shipped that takes actions a user cannot undo.
shortlistapp.coLaunched2026
Icebreakr on the Chrome Web Store
AI outreach with multi-provider fallbacks, prompt-injection hardening and an LLM-as-judge eval harness.
useicebreakr.comShipped2026
Stockpot · Agentic Commerce Hackathon
Agents trading surplus food between restaurants, settling on real payment rails. 749 tests, including an adversarial suite.
getstockpot.shopLaunched2026
NetagiriFiles
Civic transparency for ~800 Lok Sabha MPs, serving 500+ monthly visitors.
netagirifiles.fun● You're here — open to work, taking calls, shipping next.
Writing
Breaking Systems
An ongoing series on Medium and LinkedIn where I take apart the systems I work on — agent architectures, retrieval, and the distributed-systems plumbing underneath. The recurring idea: most “advanced” infra is simple once someone bothers to draw the picture properly.
Currently
Where I'm at, right now.
Status
Open to work — full-time, remote-friendly.
Looking for
AI engineer, AI agent developer, and applied-LLM roles. Go backend work too, where the agents need something reliable underneath.
Based in
Bengaluru, India (UTC+5:30).
Stack
Python · LangGraph · Pinecone · FastAPI · Pydantic · LangSmith · OpenTelemetry · Go · Postgres · Docker
Building
Shortlistapp.co — an agent that applies to jobs end to end, with paying customers.
Reply time
Same day, usually within a few hours — or just book a slot.
Let's build something that ships.
If you're hiring for an AI engineer or agent developer role — or you have an agent that works in a demo and falls over in production — pick a time. The call is on me.
Find me elsewhere.