About the company
Our client is a fast-growing software company.
The role
Raydar is recruiting for this role on behalf of our client. Design and build the core software that lets large language models carry out multi-step work reliably in production. You will shape how models are prompted and combined with conventional code, measure quality through rigorous testing, and diagnose real-world failures to make systems more robust.
What you'll do
- Develop the core execution framework for AI agents, including orchestration loops, tool usage and context handling.
- Run experiments with models and iterate on agent behavior across realistic, long-running workflows.
- Define where model decisions end and deterministic, type-checked code begins.
- Build evaluation suites on realistic test environments that are dependable enough to gate releases.
- Analyze production failures and trace each one to the right layer, then deliver systematic fixes.
- Extend browser and computer-use agents to systems that lack APIs, with strong safety, security and auditability guarantees.
- Create feedback loops and data systems that bring higher-quality real-task data into evaluation and training.
Requirements
What we're looking for
- Experience with agent frameworks or LLM systems that use tools.
- Strong Python or TypeScript skills and comfort with modern AI tooling.
- Hands-on background measuring or improving model quality, whether through testing, tuning or crafting instructions for models.
- Ability to own systems end to end and debug across the whole stack.
- A systems mindset focused on user outcomes as well as model metrics.
- Satisfaction in tracking down unpredictable production problems and converting what you learn into lasting fixes.
- At least 4 years of relevant experience.
- Ability to work in person five days a week at a startup pace.
Bonus points
- Experience building computer-use or browser-automation agents.
- Familiarity with virtualization and sandboxed execution environments, including scaling them.
- AI research experience with publications at leading conferences.
- Experience delivering systems where correctness must hold through partial failure, such as idempotency and resumability.
- Experience integrating enterprise SaaS APIs.
- Comfort working with sizable, unstructured data sources such as application logs.
- Early-stage engineering experience that included working directly with customers.
Benefits
Compensation and benefits
- Base salary: USD 200,000 to 300,000 per year
- Health insurance
- Unlimited paid time off
Location and work model
- San Francisco, CA, United States
- On-site, five days a week in office
- Full-time
Skills Required
- Experience with agent frameworks or LLM systems that use tools
- Strong Python or TypeScript skills and familiarity with modern AI tooling
- Hands-on experience measuring or improving model quality through testing, tuning, or model instruction design
- Ability to own systems end to end and debug across the whole stack
- Systems mindset focused on user outcomes and model metrics
- Ability to investigate unpredictable production problems and implement lasting fixes
- At least 4 years of relevant experience
- Ability to work in person five days per week at a startup
- Experience building computer-use or browser-automation agents
- Familiarity with virtualization and sandboxed execution environments, including scaling them
- AI research experience with publications at leading conferences
- Experience delivering systems resilient to partial failure, including idempotency and resumability
- Experience integrating enterprise SaaS APIs
- Experience working with sizable, unstructured data sources such as application logs
- Early-stage engineering experience involving direct customer interaction
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.
