A support agent that asks before it acts
customer-ops-agent: answers from a knowledge base and takes four real actions (ticket, CRM update, email, booking), every mutating one gated behind human approval.
I build AI systems that survive production.
RAG pipelines, agent workflows, and automation infrastructure. Designed, shipped, and maintained by the same engineer who runs systems like these in production every day.
Systems, not demos
A WhatsApp booking agent that cannot double-book
ConverseIQ: a multi-channel agent for a booking-heavy business that answers from the business's own records, books appointments, captures leads, and routes phone calls into the same brain.
A financial research agent for the Pakistan Stock Exchange that shows its sources
FinTex: four agents over Qdrant documents, Supabase time-series and PSX prices, fused into cited answers with a confidence score. A three-person final year project.
An ops hub where the AI can't certify its own claims
OpsRelay: calls, forwarded emails and typed notes go in; deduplicated, confidence-tagged tasks come out, behind a Slack approval gate a human has to pass before anything is sent or marked confirmed. Built for real-estate offices, agencies and dispatch desks.
The judgment is the product
Right-sized tooling
An n8n workflow when that's enough. A LangGraph backend when it isn't. You're paying for the judgment about which one your problem actually needs, not for the most impressive possible architecture.
Production mindset
Error handling, retries, evals, observability. The difference between a demo and a system is everything that happens after the happy path. That's the part I build first.
Direct
No account managers, no handoffs, no "our team will get back to you." You talk to the person who designs, builds, and ships the thing.
Three ways in
System Design Sprint
One week. Your problem in, an architecture doc and build plan out. Useful even if someone else builds it.
Fixed-Scope Build
A defined system: RAG pipeline, agent workflow, automation backbone. Scoped and quoted before work starts.
Systems Partner
A retainer to extend, maintain, and iterate on systems that have become load-bearing.
One engineer, on purpose
I'm Moosa: a Bachelor's in AI from FAST Karachi, a day job building AI automation in production, and a growing bench of independent builds. The longer version.
Bring me the problem. Leave with how I'd build it.
A 20-minute architecture call. No pitch, no slide deck. If I'm not the right person for it, I'll say so and point you somewhere better.