AI Engineer specialising in real-time speech AI, with a dual background in Software Engineering and Data Science. At Acolad I design, build, and operate Lia Live, delivering AI speech translation with sub-second latency across 30+ languages, owning features end to end from applied speech research and model evaluation to production infrastructure. Beyond speech, I work across agentic systems, LLM and RAG applications, and applied ML, turning prototypes into production. Particularly passionate about AI security: building AI that is not only capable, but secure and trustworthy.
- 🎙️ Real-time speech AI: streaming ASR, LLM translation, and TTS, with latency engineering for sub-second pipelines
- 🤖 Agentic systems & LLM apps: multi-step agents, tool and function calling, RAG and GraphRAG, agentic context-aware translation
- 🧠 Applied ML & deep learning: computer vision, sequence models (LSTM, GRU, Transformer), research to production
- ⚙️ MLOps: taking models from notebook to scalable services on Kubernetes
🔭 Currently going deeper on speech model fine-tuning, inference optimisation, and on-device deployment.
| Project | What it is | Stack |
|---|---|---|
| Claude Code Power Config | A production-ready Claude Code framework codifying a multi-agent, worktree-first workflow: project memory, path-scoped rules, auto-discoverable skills, domain subagents, and git safety hooks. | Claude Code MCP Shell |
| AuraHelpdeskGraph | Support assistant combining local LLMs, vector search, and a Neo4j knowledge graph (GraphRAG) with function calling. | Python Neo4j Ollama |
| RagDocs | Private, API-free RAG over technical documentation with Milvus vector search and local LLMs. | Python Milvus LlamaIndex |
| Aircraft Refuelling Prediction | Master's thesis with Airbus: YOLOv10 detection plus a custom SizPos-GRU sequence model predicting refuelling-port position in video. | PyTorch OpenCV YOLOv10 |
Languages
Generative AI & LLMs
Speech & Real-Time AI
ML & Data Science
AI Dev Workflow
Data & Infrastructure
Dual Master's in Computational Intelligence for Data Analytics (Cranfield University) and Data Intelligence (ISEP, Paris). At Dassault Systèmes I built scalable data workflows and NLP pipelines (topic modelling, sentiment analysis, Generative AI, RAG) for competitive and market intelligence. My Master's research with Airbus at Cranfield combined LSTM, GRU, and Transformer architectures to improve predictive accuracy for automated aircraft refuelling systems.