About the company
Our client is an artificial intelligence company.
The role
Raydar is recruiting for this role on behalf of our client. Carry out hands-on research on publicly available neural networks, testing methods for the early and refinement stages of learning, and build the measurement tools that confirm whether a change genuinely helps. The role takes a research question from first idea through to a working implementation, working closely with engineers who build production systems, and calls for a high degree of independence.
What you'll do
- Plan and carry out studies on publicly available neural networks, choosing the learning method that suits each question.
- Build systems that transform records of human work into material suitable for learning.
- Explore ways to represent workflows, context and decisions so that model performance improves.
- Create measures of output quality and dependability across lengthy, multi-step tasks.
- Set up repeatable experiment and training routines.
- Work alongside infrastructure engineers to move proven methods into live systems.
- Take a research question from first idea to a working result, owning each step.
- Investigate failed training runs and tell real gains apart from results that only look promising.
Requirements
What we're looking for
- 2 to 5 years of relevant experience.
- Practical, hands-on time with publicly available neural networks, including their foundational or refinement training phases.
- Coding proficiency proven by building research prototypes, learning infrastructure or repeatable test setups.
- Track record standing up data preparation flows for learning systems and gauging trained model quality.
- Experience designing studies independently and working out why training runs fail.
- Sound grasp of how data quality, parameter tuning, measurement and study design interact.
- Ability to turn research findings into measurable gains on extended tasks.
- Comfort working on-site five days per week.
Bonus points
- Prior work at an organization centered on applied AI is a plus.
- Work with models that learn from records of how people carry out tasks.
- Familiarity with making learning runs faster and cheaper, or with spreading computation across many machines.
- Shared research outputs or public code that others can use.
- Startup or entrepreneurial experience.
Benefits
Compensation and benefits
- Equity
- Health, dental and vision coverage
Location and work model
- New York, NY, United States
- On-site, five days per week in office
- Full-time
Skills Required
- 2 to 5 years of relevant experience
- Hands-on experience with publicly available neural networks, including foundational or refinement training phases
- Coding proficiency demonstrated through research prototypes, learning infrastructure, or repeatable test setups
- Experience building data preparation pipelines for learning systems and evaluating trained model quality
- Experience independently designing studies and investigating failed training runs
- Understanding of data quality, parameter tuning, measurement, and study design
- Ability to translate research findings into measurable improvements on extended tasks
- Willingness to work on-site five days per week
- Prior work at an applied AI organization
- Experience with models trained on records of human task execution
- Experience improving training speed and cost or distributing computation across multiple machines
- Shared research outputs or reusable public code
- Startup or entrepreneurial experience
What We Do
Raydar is a talent acquisition and business consulting firm that connects world-class and emerging talent with growing organizations. It supports companies through team development, strategic hiring, and customized growth solutions, helping clients recruit roles such as engineers, product managers, executives, legal counsel, and quantitative traders. Raydar focuses on understanding each organization’s needs, culture, and long-term goals to build high-impact teams.
