Software Engineer, Agents

About the role

We are rebuilding biotech for the AI era and this role is central to that transformation. You will own the design and execution of AI agents that automate the most time-consuming steps in scientific workflows. This includes experiment design, data capture, analysis, and complex reporting for researchers around the world. You will work at the intersection of AI engineering and scientific software, ensuring our agents are reliable, intuitive, and powerful. The team operates with a bias toward action, rapidly iterating based on real customer feedback. You will be expected to be fluent with AI tools and treat them as core components of your development stack. This role requires ownership of the full agent lifecycle from prototype to production. You will help define what great agent behavior looks like in the context of life sciences.

Key facts

What you'll do

Build end-to-end AI agents that automate scientific workflows from experiment design to data analysis and reporting.

Work directly with customers to discover use cases, gather feedback, build evaluations, and onboard scientific teams onto the platform.

Engineer across the entire technology stack, developing LLM-powered primitives and shipping intuitive user interfaces inside Benchling's applications.

Continuously improve the agent platform by contributing to frameworks, tooling, and infrastructure that accelerate future development.

Shape technical direction for AI engineering and product development as the field of AI agents matures.

Collaborate closely with product managers, scientists, and other engineers to bring ambitious ideas to life quickly.

Maintain a strong product sense, iterating rapidly based on user feedback and empirical data to improve agent performance.

Drive experimentation to discover new patterns in AI development and deployment for scientific applications.

Ensure agents are built with scientific accuracy, reliability, and usability as top priorities.

Embrace the fast-paced nature of early-stage AI agent development and navigate shifting priorities effectively.

Learn and integrate biotechnology domain knowledge to inform better agent designs and user experiences.

Champion AI fluency across the organization by demonstrating best practices in prompt engineering, tool use, and model optimization.

Participate actively in the interview process, including an AI-focused exercise or discussion to evaluate practical approaches.

Contribute to a culture of curiosity, collaboration, and rigorous evaluation in the context of AI-powered science.

Requirements

2+ years of professional software engineering experience building and maintaining production systems in a real-world environment.

Experience across the full technology stack with comfort working on backend systems using Python or similar languages.

Strong frontend skills with experience in modern frameworks such as React or equivalent for building interactive user interfaces.

Demonstrated product sense and the ability to iterate quickly to refine solutions based on user feedback and data.

Curiosity and excitement about large language models and AI agents with a desire to shape their impact on scientific research.

A collaborative mindset that enables close work with engineers, product managers, and scientists to deliver innovative solutions.

Willingness to thrive in a fast-paced environment where priorities can shift and rapid experimentation is encouraged.

Interest in learning about biotechnology is required, with no prior domain knowledge expected before joining the role.

Commitment to understanding and integrating AI tools into your daily workflow as a foundational part of the job.

Nice to have

Experience contributing to open source projects relevant to AI, language models, or scientific computing.

Background in scientific domains, data pipelines, or applications used in research or biotechnology.

Familiarity with agent frameworks, LLM optimization techniques, and evaluation methodologies for agent behaviors.

Experience working with regulated environments or data governance practices relevant to life sciences.

Understanding of user research methods and qualitative feedback integration in product development.

Practical notes

This role follows a hybrid work arrangement that prioritizes in-office collaboration.

You are expected to be in the office Monday through Friday.

The engagement is full-time based in San Francisco, California.

Candidates must be authorized to work in the United States without sponsorship for this position.

All employment is subject to Benchling's standard hiring policies and equal opportunity guidelines.