EmergingAutonomous laboratories and scientific agents

Scientific Agent Engineer

Also called AI Research Agent Engineer

A Scientific Agent Engineer builds agents that carry out research tasks and show their evidence. It matters because scientists will only use an agent whose outputs they can check and whose experiments run safely.

What they do

Builds agents that help with real research tasks, such as reading literature, proposing experiments, planning runs and interpreting results. Connects those agents to instrument APIs, lab workflow engines and experiment records, so each action and claim leaves a trace. Writes task tests with scientists, so success is defined before anything is built. Sets limits on what the agent may do on its own, and what needs a human sign-off. Reviews failures with the scientists who use the agent, and tightens the workflow, the evidence checks or the boundaries in response.

What a strong hire delivers

A working scientific agent on one real research workflow, with every claim linked to evidence a scientist can inspect. A set of task tests that scientists trust and that run whenever the agent changes. A clear record of what the agent may and may not do, including biosafety limits. A way to measure how useful scientists find it and how often it makes unsupported claims.

How you can tell they are good

Ask them to walk through an agent they built for scientific work and show how a user could check its output. Strong people talk about evidence, tests and boundaries before they talk about models. They can say where the agent failed, what they changed, and which decisions they refused to automate. Be wary of demos with no failure cases.

Why it is hard to fill

The work is being done now, but usually under other titles: automation engineer, research software engineer, computational scientist. Few people have built agents and also understand experimental design and lab safety. Candidates tend to be strong on one side and thin on the other. Job titles rarely signal the mix, so we read project history, not headlines.

Where to find them

Look in laboratory automation teams, scientific software groups and robotics teams that work with physical instruments. Experimental scientists who moved into code are a good source. So are research software engineers in universities and institutes. Biotech, materials and chemistry companies, and teams building self-driving laboratories, hold many of the people already doing this work. Some come from AI teams that have worked closely with scientists.

Indicative pay

Not enough published evidence yet to give a fair range for this role, which is common for work this new. We benchmark it for you as part of a search.

Interview questions for a Scientific Agent Engineer

  1. Tell us about an agent you built for scientific work. How did a scientist check what it produced?

    A strong answer names a specific task, the evidence trail and how scientists reviewed outputs. Listen for real user feedback and changes made because of it. Worry if the answer is only about model choice or a polished demo.

  2. How did you decide what the agent could do alone and what needed human approval?

    Good answers link autonomy to risk, reversibility and cost of error. They describe approval points and logging. Worry about blanket autonomy, or about humans approving everything without any reasoning behind it.

4 more questions, with what a strong answer covers, are in the Role Kit.

Job description you can post

We are building agents that help scientists with real research tasks, from planning experiments to interpreting results. Every output must be traceable to evidence, and every action must be safe to run. We are looking for an engineer who can connect scientific reasoning to laboratory execution and prove that it works.

What you will do

  • Design and build agents that support defined research workflows.
  • Connect agents to instrument APIs, lab workflow engines and experiment records.
  • Make every claim and action traceable to evidence.
  • Write task tests with scientists and run them as the agent changes.
  • Define limits on autonomous action, including biosafety constraints.
  • Track scientist-verified usefulness and the rate of unsupported claims.
  • Review failures with users and improve the workflow.

What you will bring

  • Built and shipped an agent or automated workflow used by scientists.
  • Hands-on experience integrating with instruments, lab software or experiment records.
  • Understanding of experimental design and how results are checked.
  • Experience orchestrating multi-step agents and testing their behaviour.
  • Clear judgement about when a human must stay in the loop.
  • Able to work closely with scientists and safety teams.

Nice to have

  • Experience in a self-driving or highly automated laboratory.
  • Background in robotics or laboratory automation.
  • Experience writing evaluation sets for scientific tasks.

Free to use and adapt. If it saves you time, a link back to this page is appreciated.

Skills that matter

  • Lab automation and biosafetyAdvancedIntegrate instruments and safe experiment execution
  • Agent orchestrationAdvancedCoordinate tool-using agent execution
  • Scientific domain expertiseWorkingApply disciplinary theory and experimental standards
  • Experimental designWorkingChoose comparisons that identify credible effects

Other roles in autonomous laboratories and scientific agents

Hiring nowLaboratory Automation EngineerA Laboratory Automation Engineer connects instruments, control software and experiment records so that experiments run the same way every time. It…EmergingClosed-Loop Experimentation ScientistA Closed-Loop Experimentation Scientist decides which experiment to run next, using models and the results already observed. They help a company…EmergingAutonomous Lab Safety EngineerAn Autonomous Lab Safety Engineer sets and tests the limits on what a model can ask a laboratory to do. Without them, an agent that plans…

Related roles in other families

Hiring nowAI Drug Discovery ScientistDrug discovery and molecular designAn AI Drug Discovery Scientist uses predictive models and experimental evidence to decide which targets and compounds are worth testing. They help…Hiring nowProtein Design ScientistDrug discovery and molecular designA Protein Design Scientist designs protein sequences and structures that meet defined functional targets, then plans the experiments to prove they…EmergingMolecular Generative Modelling ScientistDrug discovery and molecular designA molecular generative modelling scientist builds models that propose new, chemically valid compounds under clear constraints. They give a drug…EmergingAI Medicinal Chemistry SpecialistDrug discovery and molecular designAn AI Medicinal Chemistry Specialist weighs model-generated compound suggestions against real chemistry judgement and decides what gets made and…

How we wrote this

Role definitions are our own editorial work, checked against occupational classifications, standards, research and employer pages. Titles vary by company, and this is a description of the work, not a job advert.

← Autonomous laboratories and scientific agentsAll new job families

Start a search

Hiring a Scientific Agent Engineer?

Tell us the work. We find the people already doing it, whatever their job title says, and you meet them already interviewed for your role.