Agentic AI Engineer

About the role

Join a new team focused on building AI capabilities for the biotechnology industry. You will be instrumental in developing internal AI tools and company-wide agentic applications. This role is for an experienced engineer who enjoys building production systems and enabling others to use AI effectively.

Key facts

What you'll do

• Design and implement the core architecture for enterprise-level agentic AI systems, including orchestration, agent frameworks, and integration with various tools.
• Write production code for at least half of your time, establishing the infrastructure for continuous integration, testing, evaluation, and deployment of AI systems.
• Develop secure and scalable AI solutions with enterprise considerations like multi-tenancy, access controls, and human oversight.
• Create resources and provide guidance to empower other teams and power users across the company to build their own AI solutions safely.
• Collaborate with data and platform teams to ensure robust data foundations for AI applications and work with department leaders to understand and transform workflows.
• Set high standards for code quality, testing, documentation, and operational practices, while also contributing to technical hiring and mentoring.

Requirements

• A minimum of 7 years in professional software engineering, with a strong foundation in systems design for production environments.
• Practical experience building production systems that incorporate LLMs or agentic patterns, covering areas like orchestration, tool usage, state management, and observability.
• Proven ability to balance deterministic and non-deterministic computing workloads effectively in system design.
• Proficiency in at least two of the following programming languages: Python, TypeScript/Node.js, or Go, with a capacity to work across the full technology stack.
• Hands-on experience with LLM APIs (e.g., OpenAI, Anthropic), agentic frameworks (e.g., LangChain, CrewAI), RAG systems for business content, vector databases (e.g., pgvector, Pinecone), workflow automation tools (e.g., n8n, Langflow), and LLM observability/evaluation tools (e.g., LangSmith, Arize).
• A history of successfully initiating and scaling new platforms, functions, or product areas from inception.
• Experience working in environments with strict security or regulatory requirements, including a solid understanding of enterprise security principles such as encryption, access controls, and auditing.
• Ability to provide technical leadership and influence without formal authority, driving direction and raising standards.
• A product-centric approach to software development, prioritizing rapid delivery and real-world impact.
• Enjoyment in taking ownership and building foundational platform components.
• Excellent communication skills, capable of explaining technical concepts to non-technical stakeholders and translating business needs into engineering plans.

Nice to have

• Prior experience in enterprise SaaS, life sciences, or biotechnology.
• Familiarity with advanced LLM orchestration patterns and frameworks like LangGraph or MCP.
• Experience with asynchronous orchestration tools such as Temporal or Prefect for long-running workflows.
• Knowledge of compliance frameworks like SOC 2, HIPAA, or GxP as they relate to AI systems.
• Experience building internal developer platforms or large-scale internal tools.
• Direct experience coaching non-engineers on the use of AI tooling.
• Preference for candidates located in or willing to relocate to the San Francisco, CA area, with relocation assistance provided.

Skills & tools

• Python, TypeScript/Node.js, Go
• LLM APIs (OpenAI, Anthropic)
• Agentic frameworks (LangChain, CrewAI)
• RAG systems
• Vector databases (pgvector, Pinecone)
• Workflow automation (n8n, Langflow)
• LLM observability/evaluation tools (LangSmith, Arize)

Practical notes

• Visa sponsorship is not provided for this role.
• Travel is not a primary requirement.
• Benefits include equity and comprehensive health coverage.
• As part of the application process, you will be asked to complete a brief AI-focused exercise or discussion.