Expertise in application security, vulnerability research, security engineering, incident response, or cloud security.
Ability to read and review code for vulnerabilities.
Experience in identifying real bugs through various methods like code review or fuzzing.
Proven track record of building tools or workflows used by security practitioners.
Strong collaboration skills with software engineers and researchers.
Comfortable in an early-stage company environment with ambiguous challenges.
Responsibilities
Engage with customers to identify key security challenges and translate them into product priorities.
Collaborate with engineering and research teams to develop and implement security capabilities from concept to deployment.
Design realistic evaluation scenarios for codebases and infrastructure to assess security measures.
Ensure that security tasks are validated through genuine work and that patches are effective against exploits.
Analyze agent performance on security tasks to inform product improvements.
Establish trustworthiness of technical outputs by aligning them with production security standards.
Benefits
Work directly with founders in a small, technical team.
Opportunity to shape the cybersecurity roadmap.
Fast-paced work environment that encourages ownership and innovation.
In-person collaboration in Sunnyvale or San Francisco.
Full Job Description
About the role
As a Cybersecurity Product Engineer, you will decide what good security work looks like for the AI agents our customers train and evaluate. You'll work directly with frontier labs and enterprise security teams to find the security problems that matter most to them, and turn them into products and evaluations that reflect how security work is actually done.
This is a senior, hands-on role at the intersection of security, product, engineering and research. You'll help shape our cybersecurity roadmap, and what you build will be used by the teams training the most capable AI models. What you'll do
Find the real problem. Talk with customers and practitioners about how they run vulnerability management and incident response today, and turn what you learn into product priorities.
Build and ship. With engineering and research, take a defensive capability such as vulnerability discovery, patch generation or alert triage from an idea to something that runs end to end for a customer.
Make evaluations hold up. Design scenarios on real codebases and infrastructure with a realistic attack surface. Define what a correct root-cause analysis or remediation looks like, and how to verify it.
Make results trustworthy. Make sure a task can only be passed by doing the real work, and that a verifier accepts a patch only when the exploit no longer works and legitimate behavior still does.
Learn from the agents. Study how agents perform on security tasks and use what you find to improve the product.
Set the bar. Review technical approaches and outputs against one question: would a security team trust this in production?
What we're looking for
10+ years of hands-on cybersecurity experience.
Depth in at least one of: application security, vulnerability research, security engineering, incident response and forensics, or cloud security.
You read code. You can review a patch and tell whether it actually fixes the vulnerability.
You've found real bugs yourself, through code review, fuzzing or exploit development.
You've built or shaped tools, products or workflows that security practitioners used, such as scanners, detection pipelines or internal AppSec tooling.
You work well with software engineers and researchers. You can explain a security problem, what it costs, and how you'd show it's solved.
You're comfortable at an early-stage company: ambiguous problems, real ownership, moving from idea to shipped.
Nice to have
Experience working with enterprise security teams or directly with customers.
CVEs, published advisories or bug bounty findings to your name.
Familiarity with security benchmarks, CWE and CVE taxonomies, SAST/DAST, or automated exploit and patch verification.
Experience with AI agents, LLM evaluations or AI-assisted security work. This isn't required; we'll teach you the AI side.
How we work
A small, technical team. You'll work directly with the founders.
In person in Sunnyvale or San Francisco.
We move fast.
To apply
Include one example of a security tool, product, investigation or project you worked on: the problem, what you did, and how you knew it worked.