AI/ML Engineer · Biomedical Technology · RAG systems · ML & imaging pipelines
Building end-to-end machine learning systems—from data and model evaluation through APIs and deployment—with a focus on healthcare and industrial applications.
I design and implement practical ML systems with emphasis on evaluation rigor, reproducible pipelines, and deployable interfaces. Background in Biomedical Technology (B.Sc.) informs work on clinical-adjacent, signal-processing, and medical imaging problems, always framed as research and portfolio demonstrations—not medical devices.
Primary stack: Python · PyTorch · scikit-learn · LangChain · FAISS · FastAPI · Docker · Streamlit · MLflow · DICOM/NIfTI · Three.js / Next.js
| Project | Description |
|---|---|
| ECG Arrhythmia Detection | Multi-model benchmark (sklearn + 1D-CNN), stratified evaluation, FastAPI, Docker, CI |
| 3D CT Scan Explorer | DICOM/NIfTI pipeline → canonical volume → meshes → synced 3D + MPR viewer |
| LLM Document Q&A | Open-source RAG: chunking, embeddings, FAISS, Streamlit UI |
| Bio-Doc AI | Clinical-document RAG with LangChain and FAISS |
| Industrial IoT Analytics | Sensor DSP, predictive maintenance, asset integrity |
| Predictive Maintenance (Oil & Gas) | Failure classification on process-style features |
| Sales Performance Analytics | Offline analytics pipeline with SQL-ready outputs |
Full portfolio: eluan216.github.io/eluanthegenius
- Prefer comparable baselines and clear selection metrics over single-model demos
- Ship runnable repositories: synthetic data paths, tests, and CI where they add signal
- Document constraints honestly (cold starts, API keys, research-only medical disclaimers)
Open to roles and collaborations in ML engineering, applied AI, and healthcare/industrial analytics.
- GitHub: eluan216
- LinkedIn: eluanatein-oguma
- Email: ogumaeluan@gmail.com