SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Edison Scientific builds and deploys AI scientist agents to accelerate science and the development of new medicines. The company is run by scientists and engineers from leading institutions across biology, physics, chemistry, and AI.
You will join the Agent Engineering team responsible for building Edison's AI agents, particularly Kosmos, the company's AI scientist agent. In this role, you'll own the development and maintenance of production agents, prototype new capabilities with internal science and engineering teams, and shape how the platform evolves.
Key responsibilities include:
- Architect, implement, and maintain the AI agents powering Edison's platform, from prototype through production
- Work with internal science and engineering teams to explore new agent architectures, prompting strategies, tool integrations, and evaluation frameworks
- Develop reusable infrastructure such as agent skills, tool-use pipelines, and benchmarks that improve every agent on the platform
- Spend approximately 30% of time with R&D partners to understand their workflows, test solutions in real environments, and bring insights back to the team
- Translate field insights and internal research into product direction and help prioritize what to build next
You'll balance deep technical work on agents themselves—making them more capable, reliable, and useful across life sciences—with customer engagement and cross-functional collaboration.
REQUIREMENTS:
- 4+ years of professional software engineering experience with production systems that real users depend on
- Experience building LLM-powered tools or applications, including prompting, context engineering, agent architectures, and evaluation frameworks
- Strong engineering foundation in Computer Science, Software Engineering, Mathematics, Physics, Data Science, or related technical field
- Proficiency in Python and/or TypeScript with comfort picking up new tools and frameworks quickly
- Ability to work across engineering, science, and product teams
- Comfortable building from scratch, driving clarity in ambiguous situations, and wearing multiple hats
PREFERRED QUALIFICATIONS:
- Experience in life sciences, biomedical research, scientific computing, or technical R&D workflows
- Experience building agent systems, tool-use pipelines, or evaluation/benchmarking frameworks for AI applications
- Contributions to open-source scientific or AI tooling
