[Hiring] Principal Research Scientist @G2

Role Description

Buyers of business software increasingly need to know how well AI agents actually perform, not just what vendors claim. G2 is building independent, rigorous evaluations of AI agents on realistic business work, to help buyers choose with confidence and to set a trusted standard for the market.

This is the most senior applied research role on that work, and it is hands-on. You will own the research agenda for how agents get tested, make sure the results mean what we say they mean, and turn them into product features and insights G2 publishes. You will work side by side with engineers and product managers, and your work will go into what G2 ships today, not only into longer-term research.

You will also raise the bar around you. You will mentor engineers, data scientists and researchers working on evaluation, review their methodology, and make the standards you set repeatable by people who did not write them.

The field is unsettled. Methods from academia, frontier labs and industry are still being worked out. Part of the job is staying on top of that work and bringing the ideas that matter into the product. The approaches that last are the simple, general ones that still hold up as models change.

In This Role, You Will:

• Own the evaluation methodology.
• Set the standard for how G2 evaluates AI agents, defend it to skeptical audiences inside and outside the company, and keep it current as models change.
• Contribute to today's product.
• Work closely with engineers and product managers to build and improve the evaluations behind G2's current offerings. Help scope features, turn methods into production-ready components, and iterate on what customers and the business need.
• Design realistic evaluations.
• Build environments and task sets that reflect the work businesses actually give AI agents, and that stay useful as models improve.
• Develop sound measurement.
• Create scoring approaches, including automated checks, LLM-based judges and human review, that are accurate, reproducible and hard to game. Measure consistency and reliability, not just average performance.
• Validate the evaluations themselves.
• Audit tasks and graders, quantify noise, and establish when differences between agents are real.
• Represent the results.
• Present and defend evaluation findings to buyers, customers and partners, and to the research community. Be the credible voice behind the numbers when the audience has a stake in the outcome.
• Mentor and multiply.
• Manage and guide researchers and engineers on evaluation design and measurement, review their work, and write the documentation and standards that let the team run evaluations without you in the room.
• Set and pursue a research agenda.
• Choose which evaluation problems matter most and carry long-running projects through to something that works, favoring simple ideas that generalize over ones tuned to this year's models.
• Stay current.
• Follow new benchmarks, papers and evaluation practices across the research community. Translate them into product and roadmap decisions, and communicate them clearly to the team.
• Share and collaborate.
• Write methodology documentation and, where appropriate, papers and public posts. Partner with data and go-to-market teams on how results are presented and used.

Qualifications

• Experience working with evaluation design or dataset curation in multi-agentic systems.
• 5+ years building or evaluating ML systems, including hands-on work with LLMs or LLM-based agents.
• 2+ years experience managing small to medium teams preferred.
• A graduate degree (MS or PhD) in machine learning, NLP, computer science, statistics or a related field, or equivalent applied research experience.
• A track record of original ideas in machine learning, or clear improvements on existing ones, shown through first-author publications or projects you led end-to-end.
• The ability to own a research agenda: pick the problems worth solving, and carry long-running projects without being directed.
• A track record of shipping ML work in a product setting, working closely with engineers and product managers.
• Experience setting technical direction for others: mentoring researchers or engineers, reviewing methodology, and raising the quality of work you did not do yourself.
• Familiarity with how AI systems are benchmarked, and a clear view of how benchmark scores can mislead.
• Sound statistical judgment: experimental design, variance, significance and agreement between human and automated raters.
• Good judgment about trade-offs, including when a practical solution now serves customers better than a more complete one later.
• Clear writing, and the ability to explain methodology to non-specialists and to defend it to skeptical audiences.

Requirements

• Research experience in agent evaluation, user simulation, or human and AI-based judging.
• Experience with conversational, voice, retrieval-based or tool-using AI systems.
• Contributions to open-source evaluation tools or public leaderboards.
• Familiarity with business software workflows.
• A considered view on the effects of AI technology on the people and markets it reaches.
• Publications at ACL, EMNLP, NeurIPS, ICML, ICLR, COLM or similar venues.

Company Description

At G2, we are committed to creating an inclusive and diverse environment where people of every background can thrive and feel welcome. We consider applicants without regard to race, color, creed, religion, national origin, genetic information, gender identity or expression, sexual orientation, pregnancy, age, or marital, veteran, or physical or mental disability status.