Offensive Security Engineer, Replit

• Experience: 7+ years of hands-on experience in penetration testing, offensive security, or vulnerability research
• Code Fluency: You are a practitioner of whitebox testing. You can navigate large codebases and have a deep understanding of modern application architectures and secure coding pitfalls
• Cloud-Native Context: You are comfortable in a cloud-native environment. While your focus is the code, you understand how it interacts with Kubernetes, Docker, and hybrid cloud infrastructure
• Engineering Skills: Strong proficiency in Go, Python, or TypeScript. You should be capable of writing custom scripts, payloads, and proof-of-concept exploits
• Adversarial Mindset: You enjoy the "hunt" and have a proven track record of manual exploitation beyond automated scanners
• Communicator: You can translate a complex code-level exploit into a clear narrative that helps engineering teams understand risk and prioritize fixes

Desirable

• We are looking for a senior-level Offensive Security Engineer to serve as a high-impact "adversary-in-residence" for Replit’s cloud-native platform
• At Replit, security isn't just about perimeter defense; it’s about the integrity of the code that powers millions of environments
• In this role, you will lead advanced "whitebox" penetration testing engagements—diving deep into our source code to identify systemic weaknesses, logic flaws, and architectural gaps
• You will simulate sophisticated adversary tactics across our web applications, APIs, and containerized infrastructure, ensuring that our AI-integrated development environment remains the most secure place for the world’s software to live
• Lead Whitebox Penetration Testing: Execute end-to-end testing with full access to source code. You will perform manual code-level inspections to uncover complex logic flaws and authorization bypasses that automated tools miss
• Simulate Adversarial Attacks: Conduct Red and Purple team engagements across our cloud-native stack (K8s, Docker), simulating how a sophisticated actor might move from a code-level exploit to infrastructure-wide impact
• Secure AI-Enabled Systems: Perform offensive testing on LLM-backed applications and agentic AI workflows, focusing on prompt injection, data leakage, and abuse of AI-driven components
• Vulnerability Research & Chaining: Identify, exploit, and demonstrate realistic business risk by chaining vulnerabilities—from the application layer down through our internal trust boundaries
• Build Offensive Tooling: Contribute to internal security frameworks and build AI-assisted testing tools to automate the discovery of common bug classes while maintaining deep manual testing depth
• Partner with Engineering: Work closely with product teams and security architects to explain root causes, influence design guardrails, and triage high-priority findings from our Bug Bounty (HackerOne) program