Aptura is seeking doctors and medical experts to join their community for ongoing AI research projects in healthcare. The role involves leveraging medical expertise to shape AI understanding of clinical interactions and workflows, with responsibilities including data labeling, model evaluations, and collaboration with AI researchers.
Responsibilities:
- Data labeling and curation: Review clinical notes, audio, videos and other modalities, identifying relations and reasoning steps per guidelines
- LLM and model evaluations: Run structured evals (human-in-the-loop) to assess accuracy, safety, clinical relevance and clinical reasoning of models
- Prompting and prompt-engineering: Design, test, and iterate prompts to improve performance and reliability
- Fine-tuning support: Create high-quality instructions/responses and explore edge-case datasets to enhance models
- Feedback and error analysis: Identify critical/failure points, propose strategies to help mediate and document findings
- Scoring/Rubrics frameworks: Develop and refine scoring rubrics for quality assurance (e.g., accuracy, clarity, safety, completeness, guideline adherence)
- Guideline refinement: Collaborate with researchers to improve annotation standards, acceptance criteria, and evaluation protocols