
Waseem Anjum · Lahore, Pakistan · Remote
I build AI agents, MCP servers and automation pipelines, and the Next.js, React and NestJS products they run inside. Based in Lahore, working remotely for a company in Bahrain.
Waseem Anjum
Profile
About me
II''mm WWaasseeeemm AAnnjjuumm,, aa ffuullll ssttaacckk aanndd AAII aauuttoommaattiioonn eennggiinneeeerr iinn LLaahhoorree,, wwoorrkkiinngg rreemmootteellyy ffoorr aa ccoommppaannyy iinn BBaahhrraaiinn.. II bbuuiilldd tthhee wwhhoollee ppaatthh:: tthhee NNeexxtt..jjss aanndd RReeaacctt iinntteerrffaaccee,, tthhee NNeessttJJSS oorr NNooddee..jjss AAPPII,, tthhee PPoossttggrreess oorr MMoonnggooDDBB ddaattaa llaayyeerr,, aanndd tthhee AAII aaggeennttss aanndd aauuttoommaattiioonnss tthhaatt ddoo rreeaall wwoorrkk oonn ttoopp ooff iitt.. II sshhiipp ooppeenn ssoouurrccee ttoooo:: BBeenncchhlliinnee,, aa ffrreeee ttooooll ddiirreeccttoorryy tthhaatt AAII aaggeennttss ccaann ccaallll oovveerr MMCCPP;; aaggeenntt--rruulleess ffoorr AAII ccooddiinngg aaggeennttss;; aanndd PPyytthhoonn nnooddeess ffoorr CCoommffyyUUII.. II ccaarree aabboouutt cclleeaarr ssyysstteemm ddeessiiggnn,, ffaasstt ppaaggeess aanndd ccooddee tthhaatt hhoollddss uupp iinn pprroodduuccttiioonn..
Focus
AI automation + full stack
Based in
Lahore, Pakistan
Stack
TypeScript · Python
Works
Remote
What I
build
For teams and founders who need working software, not a slide deck. Hire me for one lane, or for the whole stack from prompt to production.
LLM workflows moved out of request handlers and into queues: Node.js and Python pipelines with BullMQ, retries that respect rate limits, schema-validated output and human review where it matters.
Selected
work
08 featured builds: AI agents, MCP and developer tools, and full stack products. Scroll through, or open any case study.
Journal
Latest
writing

AI Engineering
How to Add a Remote MCP Server to Cursor, Claude Code, VS Code and Claude Desktop
Tested setup for adding a remote MCP server to Cursor, Claude Code, VS Code and Claude Desktop, using a real public endpoint. Copy-paste configs, curl and Inspector checks, mcp-remote, and what tools/list costs in latency and tokens.

Backend & APIs
Server-Sent Events in NestJS: Real-Time Job Progress with @Sse (Tested)
How to stream live job progress from NestJS with @Sse and RxJS, with the patterns from my Snagbox API: ReplaySubject per job, your own event ids, comment heartbeats, and why one stream per item stalls Chrome. Every behaviour tested on NestJS 12.1.3.

AI Engineering
Building an AI Automation Pipeline in Node.js with BullMQ: Queues, Retries and Rate Limits
How to move LLM calls out of request handlers and into a BullMQ queue: idempotent jobs, Retry-After aware backoff with jitter, worker rate limits, schema-validated model output and a dead-letter queue. Tested TypeScript code.
How we
ship
One engineer, direct communication and a live preview from the first week. This is how a project with me runs, from the first call to production.
Discover
A short call and a written brief. We agree on the problem, the user, the data and what done means before any code exists.
Scope, timeline, quote
Design
Flows, the data model and the system design first, so we agree on what gets built and where AI actually helps. Decisions you can react to, not mockup theatre.
Clickable direction, tech plan
Build
Short cycles with a live preview URL from week one. You watch the product and its automations grow instead of waiting for a big reveal.
Weekly deploys, honest status
Ship & iterate
Launch is the midpoint. Monitoring, performance passes, prompt and model tuning, and fixes while real users hit the product.
Production launch, support window





























