I build AI-powered SaaS apps that actually ship to production. RAG systems, embeddable chatbots, workflow automation. MERN + FastAPI + OpenAI. London-based.

I build AI-powered SaaS applications that actually ship to production — not prototypes, not demos. Real products serving real users.
I'm a full-stack developer based in London, specialising in AI integration: RAG systems, embeddable chatbots, document intelligence platforms, and workflow automation that genuinely saves teams time.
Why I do this work:
I started in MERN engineering at HATTRICK Solutions in 2024, then joined Byte Bricks in 2025 where I shipped 3 production SaaS apps end-to-end. Along the way I noticed a pattern — most "AI projects" never make it past the demo stage because nobody handles the boring-but-critical parts: auth, error handling, deployment, billing, monitoring. I built my practice around solving exactly that gap. My focus is helping founders, ops teams, and SaaS companies move from "we should add AI" to "we have AI features customers pay for."
What I bring:
• 4+ production SaaS apps shipped end-to-end
• MERN + Python/FastAPI stack with deep LLM/RAG integration (OpenAI, ChromaDB, vector search)
• Cloud deployment on AWS with Docker and CI/CD pipelines
• Measurable outcomes: 40% support-ticket reduction, 30% faster deployments, 25% API performance gains
Recent work:
• RAG Workspace — Multi-tenant document intelligence platform with per-workspace isolation (FastAPI · ChromaDB · React · JWT)
• InboxAI — Gmail-to-Slack/Notion/ClickUp automation with AI urgency classification (Next.js · OpenAI · OAuth 2.0)
• Lecture Study — EdTech platform turning YouTube lectures into structured study sessions with AI quizzes and per-lecture Q&A (MERN · OpenAI · YouTube API)
Stack: React · Next.js · Node.js · Express · FastAPI · MongoDB · ChromaDB · OpenAI API · AWS · Docker · JWT / OAuth 2.0
Who I work with:
Early-stage founders shipping their first AI feature, ops teams tired of manual workflows, and SaaS companies adding AI to existing products. If you've prototyped something with ChatGPT and now need it built properly — that's my sweet spot.
If you need someone who can take an AI feature idea from spec to deployed production code (including the unglamorous parts like auth, billing, error handling, and observability), let's talk.
Typical response time: under 24 hours.

Work Terms

Working hours: Monday–Friday, 09:00–18:00 GMT (London). Urgent fixes on weekends by prior arrangement.
Communication: I respond to all messages within 24 hours, usually much sooner. I prefer Guru messaging for project records, plus a weekly Zoom or Google Meet call on active projects.
Payment terms: All work runs through Guru SafePay for your protection and mine. Fixed-price projects are split into milestones — typically 30% to start, 30% at mid-build, 40% on delivery. Hourly projects are billed weekly via Guru's timesheet system.
Project kickoff: Once SafePay is funded, I send a short discovery document within 24 hours covering scope, deliverables, timeline, and open questions. Build starts as soon as scope is confirmed.
Revisions: Fixed-price projects include two rounds of revisions per deliverable. Further revisions billed hourly at my standard rate.
What I need from you:
• A clear problem statement — don't worry about the technical solution, that's my job
• Access to relevant accounts (OpenAI, AWS, GitHub, etc.) when required
• Decision-maker availability for weekly check-ins
What you get from me:
• Clean, documented, production-ready code
• GitHub repo with proper commit history — yours to keep
• Deployment to your cloud account (AWS / Vercel / Render)
• Loom walkthrough on delivery
• 14 days of post-launch bug-fix support included
Timezones: I regularly work with clients across the UK, EU, US East Coast, and the Middle East. Happy to shift hours for sync calls.
Languages: English (fluent), Urdu (native).