Intern, AI Engineering at Denali Therapeutics in South San Francisco

Summary: Denali Therapeutics is a biotechnology company dedicated to developing breakthrough therapies for neurodegenerative diseases. They are seeking a motivated intern for the AI Engineering team to design and build AI-powered tools that automate workflows across the drug development lifecycle. Responsibilities: - Architect and develop internal LLM-based applications that transform complex, multi-source scientific and clinical data into structured, high-quality written outputs aligned with institutional standards - Design and implement structured prompt engineering systems that encode domain-specific knowledge, templates, and writing conventions into repeatable, auditable AI workflows - Build context-assembly pipelines that programmatically combine heterogeneous source documents, reference standards, and structured data into optimized LLM input for generation tasks - Develop rigorous evaluation frameworks that benchmark AI-generated outputs against expert-authored reference documents, using both automated metrics and human review to measure quality, accuracy, and compliance - Collaborate closely with subject matter experts to capture institutional knowledge and iteratively refine AI systems that reflect Denali’s specific standards, rather than producing generic output - Present findings and deliverables to technical and non-technical stakeholders, including a final internship presentation summarizing project outcomes, evaluation results, and strategic recommendations Required Qualifications: - Candidates pursuing a Master's or Ph.D. in Computer Science, Artificial Intelligence, Machine Learning, or a related quantitative field - Self- and results-driven, organized, resourceful, adaptable and collaborative - Bold and creative thinker with a curious nature, eager to learn about the complexities of the business of healthcare and biotech - Evident communication and analytical skills - High degree of proficiency with the MS Office Suite or Google Workspace equivalent - Demonstrable hands-on experience building applications with large language models, including structured prompt engineering, retrieval-augmented generation (RAG), context window management, or fine-tuning; evidenced through a portfolio, GitHub repositories, publications, or prior project work - Strong proficiency in Python with practical experience using LLM APIs (e.g., OpenAI, Anthropic, or open-source models via Hugging Face) and NLP libraries for text processing, generation, and evaluation - Experience designing template-based or rule-driven generation systems that encode domain-specific standards into AI outputs, and comfort working with structured and semi-structured document formats (e.g., XML, JSON, PDF, DOCX) - Experience developing evaluation methodologies for AI-generated text, including automated quality metrics and systematic comparison against human-authored reference outputs; familiarity with life sciences, clinical data, or regulated industries is strongly preferred - Builder mentality with a preference for developing custom internal solutions over relying on off-the-shelf vendor platforms; strong software engineering fundamentals including version control (Git), iterative prototyping, testing, and documentation Preferred Qualifications: - Familiarity with life sciences, clinical data, or regulated industries Required Skills: Large language models, Structured prompt engineering, Retrieval-augmented generation (RAG), Context window management, Fine-tuning LLMs, Python, LLM APIs OpenAI, LLM APIs Anthropic, LLM APIs Hugging Face, NLP libraries, Template-based generation systems, Rule-driven generation systems, Structured, semi-structured document formats XML, semi-structured document formats JSON, semi-structured document formats PDF, semi-structured document formats DOCX, Evaluation methodologies for AI-generated text, Software engineering fundamentals, Version control (Git), Analytical skills, Resourcefulness