I build the backend that powers it — and the AI
that makes it actually intelligent.
I'm emmatt, a Backend & AI Engineer who builds the infrastructure underneath intelligent products. From high-performance REST APIs and data pipelines, to RAG systems, LLM integrations, and AI-powered applications that actually work in production, not just in demos.
I sit at the intersection of solid engineering and modern AI. I care about reliability, latency, and making sure the "AI layer" doesn't collapse the moment real users show up. Based in Accra — building things that work anywhere.
An AI document intelligence platform — upload any PDF or knowledge base and query it in natural language. Built with RAG architecture, semantic vector search, and a streaming FastAPI backend. Sub-2s response times at scale.
A high-throughput event streaming API handling 50,000+ req/sec with sub-20ms p99 latency. Built with microservices, Redis pub/sub, rate limiting, and a custom backpressure system designed for zero-downtime deployments.
An AI meeting intelligence tool — real-time transcription via Whisper, automatic action item extraction, speaker diarization, and a searchable meeting memory layer. Integrated with Slack and Notion for async teams.
"Emmanuel doesn't just integrate AI, he understands it at the infrastructure level. He built our entire RAG pipeline from scratch, optimised it for production, and made the whole thing feel effortless to maintain. Rare combination of backend depth and AI fluency.
Open to freelance projects, full-time roles, and interesting problems. Reply time: usually same day.