About

I’m Pankaj Garkoti, an AI and automation engineer who builds systems that work in production — not just in demos.

I specialize in LLM-powered applications, multi-agent systems, and workflow automation. My work sits at the intersection of practical software engineering and cutting-edge AI, with a focus on reliability, scalability, and real business impact.

Key Projects

CMUX — Multi-Agent Orchestration

A production-grade system for coordinating multiple AI agents working in parallel. Features supervisor hierarchy, worker management, isolated execution environments, automatic rollback on failures, and real-time monitoring. Built to handle complex tasks that require multiple specialized agents collaborating.

ChainFactory — LLM Pipeline Framework

A framework for building, chaining, routing, and managing LLM calls at scale. Designed for production reliability with built-in error handling, retry logic, and observability. Makes it straightforward to compose complex LLM workflows from simple, testable components.

FlowPilot — Workflow Automation

An intelligent workflow automation tool that streamlines business processes with smart task routing, conditional logic, and integration with external services.

Technical Skills

Languages: Python, Go, JavaScript/TypeScript
LLMs: GPT-4, Claude, open-source models (Llama, Mistral)
AI/ML: Multi-agent systems, RAG, prompt engineering, fine-tuning, model evaluation
Infrastructure: Distributed systems, API design, data pipelines, cloud architecture
Tools: Docker, Kubernetes, PostgreSQL, Redis, FastAPI, Flask

Full Background

Download my resume (PDF) for the complete picture.

Let’s Talk

Have a project that could use AI expertise?

Get in Touch