Head of AI

About Reclaim

Reclaim Security is transforming cybersecurity with an AI-driven, business-first methodology for exposure remediation. Founded by seasoned industry experts, we enable security teams to stay ahead of emerging threats while minimizing operational disruption. At the core of our offerings is PIPE™ (Productivity Impact Prediction Engine), delivering tailored, adaptive solutions that align with business goals. With up to a 90% reduction in manual efforts, Reclaim allows teams to focus on strategic efforts and eliminates security debt through automated, efficient remediation.

About the Role

You'll own Reclaim's AI: the agentic system that discovers exposures and deploys remediations autonomously, and PIPE™, our simulation engine that predicts the business impact of every security change before it's applied. You'll join a small AI team and build from there, working directly with the CPO. You'll define the roadmap, make the architecture calls, and be hands-on enough to ship alongside your team.

What You'll Do

Agentic AI & Autonomous Remediation

• Architect and evolve the AI system that reasons, plans, and executes remediation actions across enterprise security environments
• Design the decision logic that determines what to remediate, in what order, and with what approval model
• Ensure every autonomous decision can be traced, interpreted, and clearly communicated to security teams and stakeholders
• Define and enforce the boundaries within which the agent can act autonomously — hard stops, approval gates, and safe fallback behaviors

PIPE Engine

• Contribute to the ML models that simulate whether a remediation will disrupt users or workflows before it's deployed
• Improve model accuracy, expand coverage, and build feedback loops that let production outcomes retrain the models continuously

AI Strategy & Team Building

• Define the AI roadmap in close collaboration with Product, Engineering, and Innovation
• Hire and develop the AI team, set engineering practices, and establish quality bars for AI in production
• Bring the best ideas from applied AI research into shipping features

What Makes You Successful

• Builder first — you ship. Bias for working code over polished decks. When something breaks, you debug it yourself first.
• Systems thinker — you reason about how autonomous agents behave in messy enterprise environments where a wrong decision breaks someone's workflow.
• Technically deep — you still write code, review PRs, and diagnose a misbehaving model. You lead by example.
• Independent — you own your domain, make decisions with incomplete information, iterate fast, and take full ownership of outcomes.

Requirements

Must-Have

• 6+ years building and shipping AI/ML systems in production
• Hands-on experience with agentic AI systems: autonomous agents, multi-step reasoning pipelines, LLM orchestration
• Track record improving ML model performance: you've owned a model, diagnosed why it was wrong, and made it measurably better
• Strong Python and ML engineering fundamentals (PyTorch, scikit-learn, or equivalent); production-quality code
• Explainability and guardrails experience: AI systems where decisions need to be interpretable, auditable, and policy-bounded
• Evaluation discipline: evals, benchmarks, and regression suites are part of your workflow
• Cross-functional communication: you can translate model behavior and failure modes into language PMs, engineers, and customers understand
• Startup experience: you've shipped without a platform team and thrived in ambiguity
• English fluency (professional working proficiency)

Nice to Have

• Background in cybersecurity, enterprise IT, or configuration and compliance automation
• Experience with constraint-based reasoning, safe AI, or impact prediction under uncertainty
• Exposure to multi-agent orchestration frameworks (LangGraph, AutoGen)
• Familiarity with enterprise security platforms: Microsoft 365, CrowdStrike, or SentinelOne
• Experience designing human-in-the-loop approval workflows for high-stakes autonomous systems
• Prior experience building or leading an AI function from an early stage

How to Apply

Send your resume and a brief note (3–4 paragraphs) to [email protected]. Tell us about an AI system you built and shipped: what it did, what broke, and what you'd do differently.

Reclaim Security is an equal opportunity employer.