Sr Offensive AI Security Engineer II

About the role:

Samsara is building a new role at the intersection of security engineering and applied AI. You'll be the most AI-forward technologist on our Application Security team. You'll use frontier AI models to find vulnerabilities across our cloud, software, and firmware before attackers do. You'll also evaluate each new model for security use and help build the guardrails for adopting AI safely. This role reports into our AppSec team and has a clear path to shape Samsara's AI security strategy.

This role is open to candidates residing in the US except the San Francisco Bay Metro Area, NYC Metro Area, and Washington, D.C. Metro Area.

You should apply if:

• You want to impact the industries that run our world: Your efforts will result in real-world impact – helping to keep the lights on, get food into grocery stores, reduce emissions, and most importantly, ensure workers return home safely.
• You are the architect of your own career: If you put in the work, this role won’t be your last at Samsara. We set up our employees for success and have built a culture that encourages rapid career development, countless opportunities to experiment and master your craft in a hyper growth environment.
• You’re energized by our opportunity: The vision we have to digitize large sectors of the global economy requires your full focus and best efforts to bring forth creative, ambitious ideas for our customers.
• You want to be with the best: At Samsara, we win together, celebrate together and support each other. You will be surrounded by a high-calibre team that will encourage you to do your best.

In this role, you will:

• Deliver accurate, consistent security assessments of Samsara's software, code, and firmware, evaluating their resilience against AI-driven/adversarial attacks, and implementing fixes.
• Create and translate general AI security findings into Samsara-specific fixes for our application security layer— including adapting techniques built for software to apply to hardware and embedded firmware.
• Find both AI-specific and traditional vulnerabilities (e.g., injection, IDOR) using AI-driven offensive validation.
• Build and pressure-test novel AI-enabled security tooling and workflows, then drive adoption of what proves valuable across the broader security org.
• Benchmark each new frontier model against the previous one for security tasks, and build the release criteria and guardrails for adopting it.
• Support AppSec vulnerability triage through an office-hours rotation, and help run our bug bounty program, with potential expansion to hardware.
• Partner cross-functionally with engineering, hardware, and leadership. Present findings to security and executive leadership as Samsara's go-to expert on AI-related security risk.
• Help define & drive Samsara's medium- and long-term strategy for adopting AI safely and effectively within security operations.
• Use frontier AI/LLM tools (e.g., Anthropic's and OpenAI's cybersecurity-specific programs) hands-on for both offensive validation (red-teaming our own systems) and defensive tooling — required experience integrating, securing, or red-teaming AI-enabled systems (e.g., LLMs, agentic workflows, prompt injection/adversarial testing) is essential to this role.
• Champion, role model, and embed Samsara’s cultural principles (Focus on Customer Success, Build for the Long Term, Adopt a Growth Mindset, Be Inclusive, Win as a Team) as we scale globally and across new offices

Minimum requirements for the role:

• 8+ years of experience in application security, vulnerability management, threat modeling, offensive security, or a closely related discipline, with demonstrated application fix impact and influence across a large part of an organization.
• Hands-on experience building AI or agentic workflows for security use cases.
• Experience conducting technical security assessments (e.g. offensive security or validation exercises) across both software and hardware/firmware environments.
• Experience in creating solutions for secure code with engineers using languages such as GoLang.
• Experience developing AI and agentic solutions using Github, and developing within AWS cloud.

An ideal candidate also has:

• Experience with AI-specific attack techniques (prompt injection, jailbreaks, model/agent abuse) for vulnerability management and application security.
• Experience building or validating security controls for agentic AI, RAG systems, or LLM-integrated platforms.
• Experience red-teaming LLM-enabled systems (prompt injection, agent abuse).
• Experience evaluating or benchmarking LLMs.
• Bug bounty platform experience (e.g., Bugcrowd).
• Experience in federal or FedRAMP-style environments.
• A track record of driving org-wide security strategy or presenting technical findings to senior/executive stakeholders.
• Relevant industry certifications (e.g., OSCP, GIAC, CISSP) or AWS certifications.