AI ENGINEER · AGENTIC SYSTEMS · MCP · RAG

I build AI agents that work close to the hardware.

I design agentic AI systems for the firmware and server-infrastructure world: MCP architectures, production RAG and LLM gateways. I write about AI, chips and supercomputers, and I mentor engineers trying to break into the field.

/var/log/mustafa.loguptime 4+ yrs
  1. [ OK ]agents.online LangGraph · MCP · multi-agent
  2. [ OK ]mcp.store org-wide plug-in AI integrations
  3. [ OK ]rag.production +30% employee productivity
  4. [ OK ]security.agent 15–30 min → ~30 s
  5. [ OK ]bmc.telemetry Fluent Bit → Grafana
  6. [ OK ]demos OCP Summit ’23 · SC24
  7. [ .. ]patents 5 U.S. applications · pending
  8. [ ON ]mentoring accepting mentees
01 / About

From a CS classroom in India to production AI in the US.

I started as an enterprise software engineer, came to the US for my master's in computer science, and landed in a corner of tech most AI people never see: the firmware and management layer that keeps servers alive.

That's where I learned to build AI that has to be right. My first project was the company's first RAG chatbot. It grew into a production agentic system, then a customer-facing product. Along the way I designed an MCP architecture that lets engineers plug AI agents into real enterprise and firmware APIs, and I demoed AI tooling on the floor at OCP Global Summit and Supercomputing.

Now I own an LLM gateway that routes production AI traffic across providers, and I spend my evenings writing and mentoring, because nobody handed me a map into this field and I'd like to draw one for the next person.

02 / Selected work

Things I've built and shipped.

MCPorg-wide architecture

MCP store architecture

A Model Context Protocol store that lets engineers across the organization plug in modular AI agent integrations instead of writing one-off scripts, standardizing how enterprise IT and firmware APIs connect to LLM applications.

MCPAgent toolingDeveloper platforms
+30%employee productivity

AMILiA: production agentic RAG

Architected and deployed a multi-step agentic RAG system over internal documents with zero external data uploads. Helped turn it into a customer-facing product, using RLHF feedback loops and LLM evals to improve retrieval accuracy and reduce hallucinations.

LangGraphRAGEvalsRLHF
~30 sdown from 15–30 min

Security & SBOM agent

A LangGraph multi-agent app that lets engineers ask about CVEs and firmware/software package composition in plain English, built on vulnerability-management and SBOM APIs. Later converted into an MCP server.

Supply-chain securitySBOMMCP server
Gatewayproduction LLM traffic

AI Router (AIR)

Own an LLM gateway built on LiteLLM, now an external-facing SaaS platform: multi-provider support, observability and guardrails for production-scale AI traffic.

LLM gatewayObservabilityGuardrails
OCP · SC24live demos

Conversational dashboards & BMC telemetry

Built a conversational AI tool that reads and edits Grafana dashboards through backend APIs, demoed at OCP Global Summit 2023 and Supercomputing 2024, where it helped win a major server customer. Before that, built telemetry pipelines streaming Baseboard Management Controller logs into Grafana.

Fluent BitGrafanaBMC
0external API calls

On-device Text-to-SQL

A Text-to-SQL app powered by an on-device LLM for a customer proof of concept: secure, low-latency querying of enterprise data with nothing leaving the machine.

Edge AIText-to-SQLPrivacy
03 / Patents & research

Original work on the record.

Co-inventor on five U.S. patent applications, all pending. Five more inventions are in internal review.

  1. AI-Enabled Analyses of Vulnerabilities in Software Supply ChainAppl. No. 19/636,373
  2. Agentic Pipeline for Hardware-Configuration-Driven Firmware PortingAppl. No. 19/455,400
  3. Convergence of Firmware APIs with Model Context Protocol (MCP) for AI AgentsAppl. No. 19/327,134
  4. Dynamic Classification of Geographically Distributed Nodes for Fine-Grained Management and OrchestrationAppl. No. 19/229,135
  5. Enabling System Telemetry Data Lake and Custom Data Retrieval for Analyses Using Generative AIAppl. No. 19/219,432
PUBLICATION · 2021

Classification of Various Diseases Using Machine Learning and Deep Learning Algorithms ↗

Journal of Science & Technology. Early work applying ML and deep learning to medical diagnosis, and the start of my interest in AI for healthcare.

Read the paper ↗

04 / Writing

100 days of AI, agents and chips.

A new post every day starting November 16: MCP, shipping AI to production, firmware and infrastructure, chips, supercomputers, medtech, and notes for people starting out.

Follow on Medium ↗

  • MCP Is the USB-C Moment for AI Agents. Here's What That Actually MeansOn Medium ↗
  • I Shipped an Agentic RAG System to Production. Here Are 7 Things No Tutorial Told MeOn Medium ↗
  • Every Server Has a Secret Second Computer. I Spent a Year Streaming Its ThoughtsOn Medium ↗
  • Everyone Is Building Their Own AI Chip. An AI Engineer Explains Why It Matters to YouOn Medium ↗
  • I Ran an LLM With Zero Internet Access for a Customer Demo. Here's Why It WorkedOn Medium ↗
  • What I Learned Demoing AI at the World's Biggest Supercomputing ConferenceOn Medium ↗

Trying to break into AI or hardware? Let's talk.

I mentor students and early-career engineers, especially international students and people moving from software into AI infrastructure. No fee. Bring your questions, your resume or your half-finished project.

contact@ai-khan.net

  • Your first AI agent, RAG or MCP project
  • Resume and portfolio reviews for AI roles
  • Moving from software into AI infrastructure
  • Navigating a CS master's in the US
  • Picking what to learn next, and what to skip
05 / Off the clock

Fun facts, lightly overclocked.

favorite_protocol

MCP. HTTP is a close second, and it knows it.

debugging_philosophy

Read the logs. Then read them again. The answer is always in the logs.

current_side_quest

Writing 100 posts in 100 days about AI, agents and chips.

hot_take

Tool descriptions are the most underrated prompts in AI.

will_talk_for_hours_about

Agents, chips, supercomputers, and the secret second computer inside every server.

easter_egg

There's a working terminal at the top of this page. Try sudo hire mustafa.

06 / Experience

Where I've worked and studied.