Reinforcement Learning Infrastructure (Cybersecurity)

Job Summary

The Bugcrowd RL and Reasoning Team focuses on pushing the boundaries of autonomous cybersecurity by building authentic reinforcement learning environments for foundational model companies. As a Staff Engineer, you will advance the frontier of AI Reinforcement Learning development and delivery.  You will build the infrastructure and tooling that transforms real-world vulnerability research into large-scale reinforcement learning environments used to train next-generation AI systems.

This role is unique. You will help create the training environments that teach AI systems how to hack and defend software. Your work will directly influence the capabilities of the next generation of AI models. Instead of building a single application, you will build the infrastructure that generates thousands of environments used to train frontier AI systems.

Our team works at the intersection of AI, security research, and systems engineering, building environments that allow models to learn skills such as vulnerability discovery, exploitation, and remediation. 

Essential Duties and Responsibilities 

If you enjoy building high-performance systems that power cutting-edge AI research, this role is for you.

This role focuses on building the systems that generate RL environments, not just the environments themselves. You will design pipelines that ingest software projects, analyze them with Bugcrowd’s Mayhem platform, and automatically construct training environments used by frontier AI labs including Anthropic, OpenAI, and Cohere.

The ideal candidate is a strong systems engineer who understands:

• Reinforcement learning workflows
• Building clean, reproducible Linux ML environments (containers, MCP, etc)
• System security background in binary exploitation, such as buffer overflows, fuzzing, exploitation, and x86/64.
• Experience developing applications in Python and C, with Rust a plus. 

Education, Experience, Knowledge, Skills, and Abilities

Understanding of RL training workflows used by modern LLM systems

• Experience with DevOps pipelines (e.g., github actions), reproducible builds (docker, buildkit, nix).
• Proficiency in Python and C. Other languages (especially Rust) are a plus.
• Understanding of software vulnerabilities, fuzzing, or program analysis
• Experience with build systems and large open-source codebases
• Comfort working with Linux systems and low-level debugging
• Experience working with benchmark environments (CTFs, SWE-bench, security challenges, etc.) is a plus

Working Conditions and Physical Requirements

The ideal candidate must be able to complete all physical requirements of the job with or without reasonable accommodation.

Sitting and / or standing - Must be able to remain in a stationary position 50% of the time

Carrying and / or lifting - Must be able to carry / move laptop as needed throughout the work day.

Environment - remote, work-from-home 100% of the time.

ADA Statement: 

Bugcrowd is committed to the full inclusion of all qualified individuals. In keeping with our commitment, Bugcrowd will take the steps to assure that people with disabilities are provided reasonable accommodations. Accordingly, if reasonable accommodation is required to fully participate in the job application or interview process, to perform the essential functions of the position, and/or to receive all other benefits and privileges of employment, please contact HR at ada@bugcrowd.com.

Pay Range Disclosure

At Bugcrowd, we strive for fairness, equality and to create an environment that allows our people to perform at their very best. Our compensation philosophy is to foster a collaborative community that rewards, attracts and retains the best possible talent. The provided salary details are based on US national averages and we retain the flexibility to tailor to the needs of the business.

The national estimate for the current base range for the position of $176,400 - $242,550.

This position may also be eligible to participate in a discretionary bonus program or commission plan, subject to the rules governing the program, whereby an award, if any, depends on various factors, including, without limitation, individual and organizational performance.