I am Ben Onwurah, an Applied AI Engineer specializing in production-ready AI systems that transform data, documents, and conversations into actionable intelligence and useful tools.
I design and deploy end-to-end AI pipelines — from RAG systems and content automation to multimodal voice assistants and speech intelligence tools. My work combines LLM orchestration, vector search, speech-to-text, text-to-speech, and human-in-the-loop workflows to build real-world AI products for businesses.
If you want AI systems that work, scale, and generate value, that’s exactly what I build.
- Languages & Backend: Python, Flask, FastAPI
- AI & ML: LLaMA, GPT, Gemini, Whisper, Deepgram, Pyannote, etc.
- Vector Search & RAG: Pinecone, LangChain
- Multimodal & Automation: TTS (ElevenLabs, Google Cloud, Openai), Audio Processing, Docker, Supabase, n8n
- Frontend & Interfaces: React, jQuery, HTML/CSS
- Deployment: Docker, Hugging Face Spaces, Google Cloud, AWS
Semantic search over PDFs with LLaMA 3 + Pinecone, delivering fast, grounded answers.
Recursive AI content pipeline generating human-like LinkedIn posts with multi-LLM evaluation loops.
Low-latency conversational AI with speech-to-text, LLM reasoning, and text-to-speech responses.
Speaker-aware transcripts, automated summaries, and shareable PDF reports.
Each project demonstrates full-stack AI engineering, combining system design, orchestration, safeguarding, and production-ready deployment.
.png)
.png)