← Back to Home

Hire me · Full Stack AI Engineer

Hire an AI agent developer

I build AI agents that do real work: LangGraph agents that plan and use tools, MCP servers that connect Claude and ChatGPT to your product, and CRM AI integrations that keep your data in sync. Then I ship them to production with the auth, security and monitoring they need.

Start a project →
Experience
4+ years, full stack and AI
Shipped
MCP connectors for Claude and ChatGPT
Integrations
8 CRM and data platforms
Based in
Lahore, Pakistan · remote

What I build

01

AI agent development

AI agent developer · LangGraph developer

Agents that plan multi-step work, call your tools and remember context, built with LangGraph and LangChain, with tracing so every run can be checked.

  • LangGraph planners that break a task into steps and recover from failed ones
  • Tool calling against your own APIs and databases
  • Short- and long-term memory so the agent keeps context between runs
  • Tracing and evaluation with LangSmith before anything ships
  • LangGraph
  • LangChain
  • LangSmith
  • Python
  • FastAPI
  • TypeScript

Proof

Eventas: LangGraph assistant for event operations →How I design agents →
02

MCP server development

MCP server development · MCP connectors

Model Context Protocol servers that give Claude, ChatGPT and other agents safe, scoped access to your product's data and actions.

  • Tool and resource design that models can use reliably
  • Remote MCP servers secured with OAuth2, scoped per tenant
  • Connectors for Claude and ChatGPT on top of your existing API
  • Docs and examples so your customers can switch them on themselves
  • MCP
  • TypeScript
  • Node.js
  • OAuth2 / PKCE
  • Claude
  • ChatGPT

Proof

Retell AI: Claude and ChatGPT MCP connectors →
03

CRM AI integration

CRM AI integration · AI phone agent integrations

Connect AI voice and chat agents to the CRMs your team already uses, so calls, transcripts and outcomes land in the right record with no manual work.

  • Salesforce, Zoho, GoHighLevel, Notion, Airtable, Google Workspace, Slack and Epic FHIR integrations
  • OAuth2/PKCE sign-in and HMAC-signed, fail-closed webhooks
  • Idempotent writeback of transcripts, sentiment and call outcomes
  • Zapier, Make.com and n8n apps for no-code customers
  • Node.js
  • TypeScript
  • PostgreSQL
  • Redis
  • Webhooks
  • Zapier · Make · n8n

Proof

Retell AI: multi-tenant CRM integration platform →
04

RAG and knowledge assistants

RAG developer · AI chatbot development

Assistants that answer from your own documents and data instead of guessing, with retrieval you can measure and improve.

  • Document ingestion, chunking and embeddings
  • Vector search with Pinecone, FAISS or Chroma
  • Retrieval tuned and evaluated against real questions
  • Streaming chat interfaces in Next.js and React
  • LlamaIndex
  • LangChain
  • Pinecone
  • FAISS
  • Chroma
  • Next.js

Proof

Drishti: LLM chatbot →GenAI in 50 posts series →
05

Full-stack AI products

Full stack AI engineer · AI SaaS development

The product around the AI: multi-tenant SaaS, dashboards, payments and the APIs behind them, built and run in production.

  • Next.js and React front ends
  • NestJS, Node.js or FastAPI backends with role-based access
  • Stripe, PayPal and Plaid payments and banking
  • Deployment on AWS with CI/CD

How we'll work

  1. 01

    Scope

    We agree on the one job the agent or integration must do, the systems it touches and how we'll know it works.

  2. 02

    Prototype

    A working version on your real data and APIs early, so decisions are made on results rather than slides.

  3. 03

    Production

    Auth, multi-tenant isolation, retries, monitoring and evaluation: the parts that make it safe to switch on for customers.

  4. 04

    Handover

    Documented code you own, plus support while it settles in production.

Common questions

What kind of projects do you take on?

AI agents, MCP servers, CRM and platform integrations for AI products, RAG assistants, and the full-stack SaaS around them. I work best when I can own a piece end to end, from API design to production support.

Which models and frameworks do you use?

LangGraph, LangChain and LlamaIndex for agents and retrieval, with Claude, OpenAI models or open models like Llama depending on cost, speed and privacy needs. The assistant on this site, for example, runs on open models with automatic fallbacks.

Can you work with our existing stack?

Yes. I write TypeScript (Node.js, NestJS, Next.js) and Python (FastAPI), and I've integrated with Salesforce, Zoho, GoHighLevel, Slack, Google Workspace, Epic FHIR, QuickBooks, Plaid and Stripe.

Where are you based and how do we work together?

I'm in Lahore, Pakistan (UTC+5) and work remotely with teams abroad. On Retell AI I met weekly with their Technical Director and Lead Backend Engineer to align API contracts, security and releases.

How do I get started?

Email me a few lines about what you're building and where AI fits in. I'll reply with questions or a first plan.

Have an agent in mind?

Tell me what you're building and where AI fits in. I'll reply with questions or a first plan.

Email me →Message on LinkedInResume (PDF)