We are seeking a Senior AI Engineer to lead the design and development of agent-based AI systems that transform how computational biology work is done. This is a hands-on, high-ownership role focused on building production-grade AI tools that act as collaborators for scientists - automating workflows such as data interpretation, automatic annotation, and knowledge extraction.

You will operate at the intersection of cutting-edge AI (LLMs, agents) and real-world biological applications, helping define how agentic systems integrate into scientific research.

Responsibilities:

• Lead the design and implementation of agentic AI systems for scientific workflows
• Build internal tools using LLMs, MCP, RAG pipelines, and multi-agent frameworks
• For example, develop systems for automatic annotation of biological data, literature understanding, and knowledge extraction
• Translate complex research workflows into AI-driven, semi-autonomous processes
• Take ownership from idea to production: prototype → deploy → iterate
• Work closely with computational biologists to identify and prioritize high-impact use cases
• Define best practices for agent reliability, evaluation, and performance in scientific contexts
• Contribute to the architecture of an AI-powered platform for biological discovery
  
  

Requirements:

• 5+ years of experience in software engineering, applied AI, or ML systems
• Proven track record of owning and delivering production-grade systems
• Strong Python skills and solid software engineering fundamentals
• Hands-on experience with benchmarking, evaluation, and validation of LLM/agent-based systems, including defining success metrics, running experiments, and analyzing performance tradeoffs
• Hands-on experience with LLMs and agent frameworks (LangChain, OpenAI, Anthropic, or similar)
• Experience building tools/products used by internal users or customers
• Strong problem-solving skills and ability to operate in ambiguous, fast-moving environments
• Excellent communication skills and ability to work cross-functionally