I'm Mikhail Biriuchinskii, an R&D Data Scientist and NLP specialist in Paris, with expertise in:
- 🧾 Text & documents — OCR/HTR, multilingual, historical archives
- 🧠 LLMs — Fine-tuning, prompt design, retrieval-augmented generation
- 🗣 Speech — Low-resource languages, Whisper pipelines
- 🧰 Annotation & evaluation — FAIR data, human-in-the-loop workflows
- 🛠 Deployment & tooling — Docker, FastAPI, open-access apps
I build tools at the intersection of language, AI, and data, with a focus on open-source, multimodal processing (text, speech, image), and large language models (LLMs).
I value clarity, rigor, and collaboration—especially across tech and the humanities.
You’ll find:
- 🔍 NLP demos (classification, NER, RAG, etc.)
- 📚 Tools for linguists and researchers
- 📦 Open-source contributions (TAL, OCR/HTR, LLMs)
- 🧪 Experiments with Transformers, embeddings, and vector DBs
🧭 Open to new opportunities from September 2025.