Remote work designing experimental tasks, running them through AI agents and grading the deliverables against professional standards. Suits bench scientists with 3+ years of experimental research in pharma, biotech or academia.
Overview
We’re looking for experienced bench scientists to help train AI models. You’ll design realistic experimental-biology tasks from your own practice, run them through frontier AI systems, and grade what comes back against the standard you’d hold a colleague to.
The models can already talk fluently about biology. What they can’t yet do reliably is the real work: call a plate QC go/no-go, gate a flow panel, check a cloning strategy, or read a Western blot from the raw images. Your bench judgment becomes the benchmark those models are measured against.
What you’ll actually do
- Design realistic experimental tasks from your own workflows: the scenario, the prompt, and the supporting files an agent would need (plate maps and reader output, gel and blot images, flow summaries, protocols, QC records).
- Run tasks through frontier AI agents and grade the deliverable they produce (a QC summary, experimental plan, or troubleshooting memo) against a professional standard.
- Write clear grading rubrics that specify what a correct deliverable must contain (the right controls, the right wells excluded, the right diagnosis of a failed run), and explain why a response passes or fails.
- Flag concrete failures with evidence: controls ignored, edge effects and outliers missed, wrong normalization, plausible-sounding troubleshooting the raw data contradict, or fabricated values.
- Flag concrete failures with evidence: controls ignored, edge effects and outliers missed, wrong normalization, plausible-sounding troubleshooting the raw data contradict, or fabricated values.
- Molecular and cell biology: cloning, cell culture, and cell-based assays.
- Protein biochemistry: purification, characterization, and assay development.
- Immunology and flow cytometry: panel design and gating.
- Imaging and histology, microbiology and virology.
- In vivo and preclinical work, plus lab operations and analytical sciences.
Roles this fits
Common backgrounds: Research Scientist, Molecular / Cell Biologist, Postdoctoral Fellow.
What we look for
- 3+ years designing and running experiments in a pharma, biotech, or academic research lab (counted after undergraduate study).
- Depth in at least one of: molecular biology and cloning; cell culture and cell-based assays; protein biochemistry; immunology and flow cytometry; imaging and histology; microbiology or virology; in vivo and preclinical work.
- A hands-on practitioner who designs and troubleshoots experiments, not only executes SOPs: you can explain what each control is for and diagnose why a run failed.
- You generate and QC primary data, and can judge from raw instrument output whether a run is usable.
- Master’s or PhD (or current PhD candidate) in biology or a related field, completed in the U.S., Canada, Europe, or the UK.
- Clear written English, comfort with ambiguity, and familiarity with AI/LLM tools like Claude or ChatGPT.
- Based in the United States, Canada, or the UK (Ireland and Australia may also be accepted).
How it works
Apply
Qualify
Work & get paid
Compensation
Up to $40 – $125+/hr depending on task difficulty and specialization. Many contributors add $10k–$100k+ a year; some make it their full-time income.
About DataAnnotation
DataAnnotation is where 100k+ experts train the world’s leading AI models. $150M+ paid to contributors to date, and the average contributor stays 5+ years. Flexible, remote, and always project-available.
Apply now
Listing from DataAnnotation's job board. Details can change; the platform's page is the final word.
