Open to remote · AI / Agent Engineer · UTC+5:30

I build agents
that do the work.

SUMANTH KUMAR M — 8 YEARS, SOFTWARE ENGINEERING

I design, ship, and operate production LLM agent systems — autonomous software running 24/7, not prototypes. I built and run a self-hosted multi-agent platform, and open-sourced a multi-agent code reviewer with a measured evaluation benchmark.

Ask my AI live
Hi — I'm Sumanth's assistant, and I'm an agent he built. Ask about his agent work, projects, skills, or fit for a role.
01 — Selected work

Things I've built & run.

A measured open-source agent, a production multi-agent platform, and 15+ shipped apps.

Open source · live demo

Quorum — multi-agent code reviewer

A dispatcher routes a diff to specialist LLM agents — security, performance, correctness, tests, style — in parallel; a synthesizer consolidates them into one verdict. Provider-agnostic (Claude / OpenAI / Ollama) with native structured output, a CI-gated detection-eval benchmark (recall / false-alarm scored), config-driven custom agents, and zero runtime dependencies. The demo runs the whole engine in your browser via WebAssembly.

Production · 24/7

Multi-agent ops platform

An LLM dispatcher routing tasks to specialist agents (triage, content, code execution, strategy), with a realtime task board (REST + SSE), per-task model routing, retries, heartbeats, idempotency, and agents that self-report via API.

Autonomous

Crash-triage & strategy agents

One agent watches production crash signals and drafts/routes fixes automatically; another scores candidate actions by expected value and surfaces one ranked recommendation per day.

Meta

This site's "Ask my AI"

The agent above is grounded in my background and scoped to answer only about me — a serverless LLM behind a thin proxy, no API key exposed. The portfolio of an agent engineer should itself be an agent.

Shipped

Cross-platform app portfolio

15+ production apps (Flutter / Android / iOS) with Firebase, CI/CD, 15-language localization, and on-device + cloud AI — 100K+ combined installs at 99.5% crash-free.

02 — Toolkit

What I work with.

Deep in LLM / agent engineering, broad across the stack.

AI / LLM

Claude · OpenAI · Ollama orchestration · multi-agent systems · tool use & function calling · structured output · prompt engineering & caching · RAG · agent memory · MCP · model routing · evals (golden-dataset, LLM-as-judge)

Languages

Python · Dart · JavaScript / TypeScript · Kotlin · Java · Swift

Backend & infra

Flask · REST · Server-Sent Events · SQLite · Firebase · Linux · systemd · cron · CI/CD · Cloudflare Workers

Mobile

React Native · Flutter · native Android (Jetpack Compose, MVVM, Coroutines) · iOS

03 — Track record

Experience.

Software Engineer  ·  Unify Technologies
FEB 2024 — PRESENT
Senior Software Engineer  ·  Robosoft Technologies
DEC 2022 — JAN 2024
Senior Software Engineer  ·  LoadShare Networks
SEP 2021 — OCT 2022
Software Engineer  ·  Robosoft Technologies
JUL 2020 — JUL 2021
Software Engineer  ·  Advanced Voice Research Labs
JUN 2018 — JUN 2020