About You
We’re seeking a senior technical leader whose background combines substantial AI/ML research expertise with hands-on engineering experience. You’ll help shape the architecture and technical direction of our intelligent systems. The remit extends beyond implementation as an individual contributor: you’ll uncover unanswered questions, develop experiments to investigate them, and take ownership of AI development from initial research through production.
Your experience should combine research depth, typically gained through a PhD or equivalent industry research, with substantial work making AI systems dependable on complex, real-world production data. You’ve created evaluation frameworks, investigated LLM failure modes, and learned to distinguish impressive benchmark results from meaningful evidence of reliability. You bring informed perspectives on measuring AI performance and deploying systems safely in environments where the consequences matter.
AI is central to both our products and our working practices. Across research, engineering, and operations, we make extensive use of AI tools to accelerate execution and support clearer thinking. Everyone is expected to adopt this approach. At a senior level, that means demonstrating how AI can amplify informed judgment without replacing it.
Healthcare offers both extensive data and significant responsibility. We’re looking for someone who understands the importance of both.
Your Responsibilities
AI Strategy & Research
● Own the technical strategy for AI/ML across our clinical intelligence products, covering information extraction, summarization, reasoning, and retrieval. Carry that strategy through to implementation.
● Assess weaknesses in existing systems, including inconsistent performance, missing coverage, and evaluation methods that overlook failures encountered in practice.
● Follow developments in LLMs, clinical NLP, retrieval-augmented generation, RLHF/RLAIF, agents, and structured prediction, bringing relevant advances into production.
● Establish how we evaluate and benchmark models, prioritizing measures relevant to patient care rather than metrics selected simply because they are straightforward to collect.
Production Systems & Engineering
● Develop and enhance production pipelines for clinical entity extraction, summarization, chart abstraction, and structured inference from unstructured clinical text.
● Shape and refine our use of RAG, prompt engineering, fine-tuning, and model selection, applying rigorous evaluation throughout.
● Oversee secure AI deployment that complies with HIPAA and SOC 2, focusing on accuracy, auditability, and the risks associated with hallucinations.
Technical Leadership & Collaboration
● Provide technical guidance on AI decisions, helping the team determine which approaches to explore, diagnose problems, and assess whether models are ready for release.
● Partner directly with clinical subject matter experts to understand how outputs support clinical workflows, where they fall short, and how quality should be defined in practice.
● Mentor engineers and strengthen the team’s understanding of AI/ML best practices, including effective use of AI tools in everyday technical work.
Required Experience & Skills
Essential Requirements
● Substantial expertise in modern AI/ML, LLMs, NLP, information extraction, or a closely related discipline, supported by direct work with leading approaches rather than theoretical familiarity alone.
● A PhD in ML, NLP, computer science, or a related discipline, OR comparable research depth evidenced by publications, substantial open-source contributions, or impactful AI work in industry.
● Experience creating and owning evaluation frameworks, including defining what to measure, accounting for Goodhart’s Law, and identifying regressions, rather than only executing existing evaluations.
● Demonstrated experience progressing AI systems from research or prototypes into dependable production deployments, with a clear understanding of their limitations and the causes of failure.
● At least 8 years of software engineering experience, predominantly focused on AI/ML systems.
Technical Expertise
● Fluency in Python, either PyTorch or JAX, and the core ML tooling used for training, fine-tuning, inference, and quantization.
● Practical experience working with LLMs through prompting strategies, RAG architectures, RLHF/preference tuning, instruction tuning, or model distillation.
● Strong data capabilities spanning SQL, pipeline design, annotation tools, and dataset curation, alongside an understanding that data quality underpins much of AI performance.
● Experience using NLP libraries such as Hugging Face and spaCy, together with production model-serving infrastructure such as vLLM, TGI, or equivalent tools.
● Skilled use of AI-assisted development tools such as Claude Code, Cursor, Copilot, or similar. You actively guide and scrutinize these tools rather than accepting their responses without challenge. You understand when to explore further, when to reject an output, and how to organize tasks to obtain useful results without overlooking subtle errors.
Working Approach
● An AI-first approach grounded in practical understanding of where AI provides meaningful benefits and where it does not. This should be reflected in your coding, research, and daily problem-solving.
● The judgment to recognize when a model should not be released, balanced with the pragmatism to deploy imperfect systems where risks are understood and contained.
● Comfort navigating open-ended research questions, combined with the discipline needed to turn uncertain findings into production systems.
● Genuine interest in the clinical domain and recognition of the responsibility involved in developing high-stakes AI.
Additional Experience
● Familiarity with clinical NLP, healthcare data standards such as HL7 and FHIR, ICD/CPT coding, or previous healthcare AI work.
● Experience in structured prediction, building medical knowledge graphs, or extracting information from clinical data.
● Publications or applied research contributions covering NLP, LLMs, clinical AI, or evaluation methods.
Technology Environment
Python · PyTorch · Hugging Face · PostgreSQL · BigQuery · dbt · FastAPI · Google Cloud Platform · LLM APIs (OpenAI, Anthropic, etc.) · Jupyter / Streamlit