I’m Muhammad Noman Ashraf, a full stack developer and AI automation engineer. I build web experiences, AI agents, and workflows that make work easier.
A selection of AI systems, connected workflows, and applications. Explore the approach, tools, and code behind each build.
01A multi-stage autonomous LangGraph agent that automates the job application pipeline. Scrapes job boards, executes vector cosine similarity matching against candidate memory, auto-drafts personalized cover letters, and dispatches rich Slack notifications.
02Autonomous multi-agent customer support workflow. Ingests incoming tickets, classifies intent and urgency via LLMs, retrieves documentation from Vector DB, drafts empathetic responses, and routes escalations to Slack.
03Coordinating six specialized AI agents to automate medical literature reviews. Fetches PubMed & Semantic Scholar data, validates cross-references, maps citations, and generates clinical research summaries.
04AI-powered image recognition and contextual analysis mobile application. Users scan any object or document to receive instant classification, real-time AI conversational Q&A, and smart gallery categorization.
Showing 4 of 15 projects
From a focused integration to a complete product, I bring the interface, logic, and automation together.
Turn complex research and decision workflows into connected, purpose-built AI systems.
Discuss a projectConnect your tools, reduce repetitive work, and keep your business moving with custom integrations.
Discuss a projectResponsive interfaces and dependable APIs, built together for a smooth end-to-end experience.
Discuss a projectThoughtful cross-platform applications that make everyday tasks feel simple.
Discuss a projectI’m Noman, a developer based in Pakistan, working across full stack development, AI systems, and business automation.
I enjoy connecting the dots between a good idea and a useful product: shaping the interface, building the API, and automating what happens behind the scenes.
React interfaces, Node.js services, and API integrations.
Agent orchestration, retrieval workflows, and n8n integrations.
Flutter applications with Firebase and connected AI features.
For years, AI meant a single prompt and a single response. Multi-agent systems powered by AutoGen are rewriting those rules — and the possibilities are staggering.
Read the article Tutorial · 7 MIN READBoth frameworks promise to simplify LLM development, but they solve fundamentally different problems. Here is how to choose the right tool for your workflow.
Read the article Tutorial · 8 MIN READCrewAI brings multi-agent orchestration within reach of every Python developer. Here is a practical guide to building your first intelligent crew from scratch.
Read the articleTell me what you have in mind. Let’s find out how I can help.
Start a conversation