|
Описание: |
Remote
Full-time
We’re seeking for a self-driven, proactive and startap-minded AI Engineer to join a German healthcare project — an AI medical “brain” for doctors and medical professionals that makes billing faster, more accurate, and fully compliant. The platform leverages state-of-the-art language models to interpret clinical notes, recommend CPT, ICD-10, and HCPCS codes, and check compliance against payer and regulatory rules.
What You’ll Do
* Build and fine-tune large language models (LLMs) for medical text understanding
* Implement retrieval-augmented generation (RAG) pipelines with vector databases
* Train and evaluate models on real clinical notes, CPT, ICD-10, and HCPCS coding data
* Incorporate logic for payer-specific and regulatory coding rules (bundling, modifiers, and regional/state variations)
* Optimize inference for speed and accuracy (quantization, distillation, GPU tuning)
* Collaborate with product + medical experts to turn research into production features
* Establish scalable workflows and documentation so the team can quickly adapt to updated LLMs, datasets, and payer/regulatory requirements
What We’re Looking For
* Strong experience in NLP/LLMs (PyTorch, Hugging Face, or equivalent)
* Background in retrieval systems, vector databases, and RAG pipelines
* Proven ability to train and fine-tune models on specialized datasets
* Ability to design processes for precision, accuracy, and rigorous benchmarking to ensure models meet compliance and real-world medical billing standards.
Nice to have
* Understanding of healthcare coding standards (CPT, ICD-10, HCPCS)
* Must have past projects we can review (GitHub, online demos, or similar)
* Bonus: experience with model optimization for local inference (quantization, GGUF, ONNX, etc.)
Why Join
* Tackle Hard Problems: Work on cutting-edge challenges at the intersection of AI, healthcare, and compliance — where every improvement directly reduces denied claims and improves patient care.
* Shape the Product: Your expertise won’t be a small cog in the machine — you’ll own critical decisions in model design, fine-tuning, and RAG architecture.
* High-Velocity Startup: Move fast, experiment boldly, and see your work ship to production in weeks, not quarters.
* Mission-Driven Team: Join people who care deeply about fixing one of the most painful problems in U.S. healthcare with technology that’s practical, ethical, and transformative.
Looking forward to your replies!
Відгукнутись на вакансію |
