NNouman Naseer
Software Architect · Austin, TX

I design systems that scale, and build with AI agents.

I'm Nouman Syed Naseer, a software architect with 8+ years building distributed systems, cloud-native platforms and high-availability services. Today I'm focused on putting generative AI, LLMs and multi-agent systems to work on real problems.

View my work Explore the Playground →
8+Years engineering
99.9%Platform uptime
1M+Events / day
0Agent builds
01 · About

What I do

I've spent my career moving teams from monoliths to microservices, making platforms faster and more reliable, and building the automation around them. Lately I've been pairing that with AI: chatbots that speed up operations, and agents that can build whole apps on their own.

Distributed systems

Microservices, event-driven and domain-driven architecture, built for high availability and clear ownership.

Cloud-native platforms

Automation platforms on IBM Cloud and AWS, with CI/CD, caching, monitoring and disaster-recovery planning built in.

Agentic AI

LLM apps, RAG and multi-agent systems with LangChain, LangGraph, CrewAI and AutoGen, connected through MCP and A2A.

02 · Experience

Where I've worked

From backend developer to software architect, working across healthcare, enterprise cloud and AI.

NowJun 2022 – Present

Software Architect · IBM

  • Architected a cloud automation platform with 99.9% uptime on Cloudant, Event Streams, MQ and Code Engine.
  • Built a GenAI operations chatbot with IBM Granite and LangChain that streamlined workflows and customer case handling.
  • Created a FastAPI Slack bot for incident management that cut response times by 40%.
  • Brought API latency from 800 ms to 150 ms with Redis caching, and made deployments 60% faster by moving legacy services to microservices.
  • Set up CI/CD with automated testing and designed the disaster-recovery strategy (RPO 15 min, RTO 1 hr).
IBM CloudGenAILangChainFastAPIRedis
Jun 2020 – Jun 2022

Software Development Engineer · Optum

  • Architected a Java microservices platform handling 50K+ daily transactions at 99.95% availability, with a React/TypeScript front end for 10K+ concurrent users.
  • Designed event-driven systems on Kafka processing 1M+ events a day in real time.
  • Brought key database queries from 2.3 s to 400 ms with pooling and indexing.
  • Set up monitoring and alerting that made issue resolution 38% faster.
JavaSpring BootKafkaReactTypeScript
Dec 2017 – Dec 2020

Backend Developer · Techlogix

  • Moved a monolith to Java/Spring Boot microservices, cutting report generation time by 30%.
  • Built a React reporting dashboard with real-time charts that halved manual effort.
  • Delivered OAuth 2.0 REST integrations with 5 external systems, and raised test coverage from 65% to 92%.
Spring BootActiveMQOAuth 2.0
03 · Skills

Toolbox

The languages, platforms and patterns I reach for most.

Education
B.S. Computer Science
National University of Computer and Emerging Sciences (FAST-NU), Lahore · 2013 – 2017
05 · Contact

Let's build something reliable, and a little intelligent.

Open to conversations about architecture, platform engineering and applied AI. The quickest way to reach me is email or LinkedIn.

Email me LinkedIn ↗ GitHub ↗
Playground

Apps built by AI agents.

A lab of games, apps, docs and designs that AI agents built end-to-end. Open one to try it live, then read the docs for controls, rules and how it was made. Each card shows which agent built it.

Growing the playground

How projects get here

Each project lives inside this site as its own page, with a card, a live "Try it" view and its own documentation. Any AI agent can be the builder.

01

An agent builds it

A game, tool or app made by an AI agent such as Claude. Anything that runs as a web page works best.

02

It's added here

The project is copied into the Playground, screenshotted, documented and tagged with the agent that built it.

03

Reviewed, then live

Every change gets a preview link first. Once approved, it goes live on nouman.ai.

Add my latest project to the Playground on nouman.ai, with docs and a Try it view.