Data Scientist & ML Engineer with 5+ years taking models from notebook to production — named-entity recognition, retrieval-augmented generation (RAG), LLM agents, real-time IoT anomaly detection, and large-scale MLOps. M.Sc. in Statistics (Data Science), first in a cohort of 10, with peer-reviewed research in EEG deep learning and medical image segmentation. I build systems that move real metrics: revenue, retention, and time-to-production.
- 🔭 Currently: mentoring at Great Learning, industrial research on explainable anomaly detection for 69-machine fleets (SPC + TCN-autoencoder + SHAP, Kafka production deployment).
- 🌱 Focus: production LLM/RAG systems, MLOps & monitoring, time-series & anomaly detection, uncertainty quantification.
- 💬 Ask me about: shipping ML to production, RAG pipelines, statistical inference at scale, calibration of predictive intervals.
- 🌍 Open to Data Scientist / ML Engineer roles — remote or relocation (EU · Canada · Australia; visa sponsorship).
| Project | What it is |
|---|---|
| Interactive Portfolio + RAG Chatbot · Live ▶ | Multi-page Streamlit portfolio with a LlamaIndex + OpenAI chatbot that answers recruiter questions over my résumé. |
| EEG Deep Learning for ADHD | CNN / custom ResNet on EEG time-frequency images — ~98.6% subject-level accuracy across 121 children. Published in Cognitive Computation (2024). |
| Mortality Calibration Under Shift | Pre-registered audit of whether mortality prediction intervals hold nominal coverage across the COVID-19 structural break. Ten forecasting families × seven uncertainty mechanisms, 50 crossed cells. |
| LLM Zoomcamp Capstone | "IT Group Assistant" RAG app: FastAPI + OpenAI, Postgres logging, Grafana monitoring, Dockerized, with LLM-as-judge evaluation. |
| MLOps Zoomcamp Course | End-to-end MLOps: experiment tracking, orchestration, deployment, and monitoring. |
| Mathematical Statistics with R | Code companion to my textbook — estimation, hypothesis testing, and distribution theory via Monte-Carlo simulation in R. |
Peer-reviewed journal articles
- M. Shafiei Neyestanak, H. Jahani, M. Khodarahmi, J. Zahiri, M. Hosseini, A. Fatoorchi, M. S. Yekaninejad. A quantitative comparison between focal loss and binary cross-entropy loss in brain tumor auto-segmentation using U-Net. Journal of Biostatistics and Epidemiology, 11(1), 2025. doi:10.18502/jbe.v11i1.19315
- H. Jahani, A. A. Safaei. Efficient deep learning approach for diagnosis of attention-deficit/hyperactivity disorder in children based on EEG signals. Cognitive Computation, 16(5), 2315–2330, 2024. doi:10.1007/s12559-024-10302-3
Book chapter
- H. Jahani, A. A. Safaei. Neural signals processing using deep learning for diagnosis of cognitive disorders. In Handbook of Neural Engineering: Signal Processing Strategies, Vol. 1, Elsevier, 2025. doi:10.1016/B978-0-323-95437-2.00005-7
Textbook
- H. Jahani. Mathematical Statistics with R. Allameh Tabataba'i University Publishing Center, Tehran, 2020.
- 🥇 Gold medal, Iran Statistics Competition (2019) — highest score in the competition's history.
- 🥈 Silver medal, Iran Statistics Olympiad (2019, 2020).
- 🥈 Second place, Iran National Master's Entrance Examination in Statistics (2020).
- 🥇 First place (two events), Play with Real Data competition among 60 teams, AGNA.co (2023).
-FCC624?style=for-the-badge&logo=linux&logoColor=black){kind=icon}
