Forward Deployed Engineer · Distributed Systems · AI Reliability
I build systems where being wrong isn't an option. Then I point that discipline at AI — and translate it for the people who have to trust it.
I'm Vipul — a backend engineer on a financial core system, where a mismatched record isn't a bug, it's an audit finding. I make distributed systems provably correct, build open-source tooling that tells you whether an AI agent actually holds up in production, and explain both to the people who don't write code but have to sign off on them.
What I actually do
Three things, and they reinforce each other.
Correctness under load
I work on a financial core serving 100M+ accounts, where migration and reconciliation have to be right the first time. I delivered 90M+ records at 99.9%+ validated accuracy, rebuilt entity loading for a 60–70% throughput gain, and wrote reconciliation that survives out-of-order events instead of silently mislinking accounts.
Does your agent actually work?
I build open-source tooling that answers the question a demo can't. Agent Reliability Workbench lets any team plug in their LLM agent, build evals from a template, and benchmark it on reliability, latency, and cost — with a full trace behind every run.
Explaining it to whoever signs off
Correctness only counts if the people who don't write code trust it. I've defended migration and reconciliation results to compliance, audit, and tech-lead reviewers — turning consistency and event-ordering guarantees into terms a non-engineer can approve. It's the part I like most.
Featured project
Agent Reliability Workbench
Benchmarking for tool-using LLM agents
A platform where anyone registers their agent and tests it against real failure modes — the ones that break systems in production, not toy prompts.
- A 12-case benchmark porting real distributed-systems failures: duplicate-key retries, DLQ poison messages, config drift, rate-limit storms, unsafe remediation.
- A typed Agent-Under-Test contract + HTTP JSON adapter — register any external agent and run it against the same suite as the built-in reference agent.
- FastAPI + PostgreSQL services for runs, trace persistence, eval execution, and comparison APIs (Pydantic, SQLAlchemy, Alembic).
- A full trace per run — schema and guardrail checks, latency, tokens, cost, model/prompt version — so agents compare on reliability, not vibes.
Experience
Where the numbers come from.
Software Engineer, Enterprise Microservices
On a 3-engineer team modernizing a legacy core account system for 100M+ accounts. I owned the Contact and Authorized User pipelines end-to-end — 700M+ records through parsing, business rules, PostgreSQL, and publishing to 4+ downstream services — and delivered migration and reconciliation at 99.9%+ validated accuracy. I also resolved a production incident where audit-field-only diffs were looping legacy-sync events and firing false downstream updates.
Python Developer Intern
Built probabilistic modeling workflows (PyMC3, ArviZ) over large financial datasets and integrated FINRA API data into quarterly projection pipelines.
B.S. Computer Science
Foundations for the systems work I do now — plus the years of explaining technical things to people who don't share the vocabulary.
Open to work
Forward Deployed & AI Solutions Engineering roles.
If you're deploying frontier models into someone's production system and need an engineer who can build it and explain it to the people who have to trust it — let's talk.