Senior Frontier Agents Engineer (Forward Deployed Engineering)

Who are we? Cohere is the leading security-first enterprise AI company. We build cutting-edge foundation AI models and end-to-end products that are designed to solve real-world business problems. We’re training and deploying frontier models for enterprises who are building AI systems. We believe that our work is instrumental to the widespread adoption of AI and we are looking for folks that want to be part of that. We obsess over what we build. Each one of us is responsible for contributing to increasing the capabilities of our models and the value they drive for our customers. Cohere is a team of researchers, engineers, designers, and more, who are all passionate about their craft. We are a global technology company headquartered in Toronto with key offices in London, New York City, San Francisco, Montreal, Paris, Berlin and Seoul. Join us! Role Overview Ready to build AI agents that revolutionize enterprise security? The North Security team is looking for a Senior Software Engineer to create autonomous systems that handle the complex, critical tasks security practitioners face every day. Key Responsibilities As a Senior Software Engineer focused on agentic security, you'll build intelligent systems that empower security teams. Your responsibilities will include: Build autonomous security agents that perform alert triage, secure code reviews, threat modeling, and vulnerability assessment Develop agent orchestration systems that help security practitioners automate repetitive security tasks Design and implement agent security controls to ensure our AI agents can safely interact with sensitive enterprise data Create the security infrastructure that allows agents to operate in high-trust environments with stringent deployment requirements Ship production-ready agent capabilities that run efficiently in resource-constrained environments Research and implement novel security patterns for AI agent systems, sometimes requiring custom solutions where standard libraries fal

Who are we? Cohere is the leading security-first enterprise AI company. We build cutting-edge foundation AI models and end-to-end products that are designed to solve real-world business problems. We’re training and deploying frontier models for enterprises who are building AI systems. We believe that our work is instrumental to the widespread adoption of AI and we are looking for folks that want to be part of that. We obsess over what we build. Each one of us is responsible for contributing to increasing the capabilities of our models and the value they drive for our customers. Cohere is a team of researchers, engineers, designers, and more, who are all passionate about their craft. We are a global technology company headquartered in Toronto with key offices in London, New York City, San Francisco, Montreal, Paris, Berlin and Seoul. Join us! Why This Role Is Different This is not a typical “Applied Scientist” or “ML Engineer” role. As a Member of Technical Staff, Applied ML, you will: Work directly with enterprise customers on problems that push LLMs to their limits. You’ll rapidly understand customer domains, design custom LLM solutions, and deliver production-ready models that solve high-value, real-world problems. Train and customize frontier models — not just use APIs. You’ll leverage Cohere’s full stack: CPT, post-training, retrieval + agent integrations, model evaluations, and SOTA modeling techniques. Influence the capabilities of Cohere’s foundation models. Techniques, datasets, evaluations, and insights you develop for customers will directly shape the next generation of Cohere’s frontier models. Operate with an early-startup level of ownership inside a frontier-model company. This role combines the breadth of an early-stage CTO with the infrastructure and scale of a deep-learning lab. Wear multiple hats, set a high technical bar, and define what Applied ML at Cohere becomes. Few roles in the industry combine application, research, customer-facing engineeri

About the role We’re looking for an engineering manager to lead a team building software systems that detect and prevent harmful misuse of frontier AI models—before incidents occur. This is a builder’s role: you’ll lead engineers shipping production services, detection pipelines, and mitigation mechanisms that protect frontier model integrity and reduce high-severity misuse risk. While this work intersects with frontier model development, security and risk, we’re explicitly seeking someone with a software engineering foundation who is comfortable building reliable systems that can operate at billions of users scale. In this role you will: Lead a team of software engineers building detection + mitigation systems for frontier model misuse, with an emphasis on model IP protection / distillation detection and emerging risk surfaces from autonomous agents. Set the technical roadmap and execution strategy: prioritize, design, ship, iterate, measure impact. Build production systems: services, pipelines, tooling, instrumentation, and automation that scale with frontier model usage. Partner deeply with Research and Product to translate evolving model capabilities into concrete tests, signals, and mitigations that can be deployed at scale. Drive strong engineering fundamentals: architecture, reliability, monitoring, performance, and operational excellence. Hire and grow an exceptional team across backend, data systems, and applied ML engineering domains as needed. Anticipate what breaks at scale as agentic workflows become more capable. You might thrive in this role if you: Experience building systems in adversarial, fast-evolving environments Are comfortable with ambiguity and novelty Have experience adjacent to security (e.g., abuse prevention, fraud, integrity, platform defense, auth/identity, malware/spam, adversarial environments) Communicate clearly and build trust quickly with senior stakeholders—pragmatic, collaborative, and calm under scrutiny. Significant experience

About the team Models are becoming increasingly capable—moving from tools that assist humans to agents that can plan, execute, and adapt in the real world. Mitigating the frontier risks resulting from these capabilities is paramount to OpenAI’s ability to continue deploying models safely. The Preparedness team is dedicated to addressing these critical risks. Our work includes: Measurement. Monitoring and predicting the evolving capabilities of frontier AI systems. Mitigation. Keeping misuse and misalignment safeguards, alignment tools, and security measures on track to adequately address extreme threats that might arise in the future. Coordination. Setting mitigation targets by maintaining OpenAI’s preparedness framework , and partnering with other staff to achieve these targets. This is urgent, fast-paced work that has far-reaching implications for the company and for society. About the role As we push toward AGI, cybersecurity becomes one of the most important and urgent frontiers: the same systems that can accelerate productivity can also accelerate exploitation. As a Researcher for cybersecurity risks, you will help design and implement an end-to-end mitigation stack to reduce severe cyber misuse across OpenAI’s products. This role requires strong technical depth and close cross-functional collaboration to ensure safeguards are enforceable, scalable, and effective. You’ll contribute directly to building protections that remain robust as products, model capabilities, and attacker behaviors evolve. In this role, you will: Design and implement mitigation components for model-enabled cybersecurity misuse—spanning prevention, monitoring, detection, and enforcement—under the guidance of senior technical and risk leadership. Integrate safeguards across product surfaces in partnership with product and engineering teams, helping ensure protections are consistent, low-latency, and scale with usage and new model capabilities. Evaluate technical trade-offs within the cybers

About the Team Critical Harm Operations sits within User Safety & Risk Operations and builds enforcement systems for the most severe user-safety risks—including Frontier Risk and Material Harm—that are accurate, fast, defensible, and scalable. The Cyber vertical turns policy into reviewer standards, calibrated judgment, quality systems, escalation paths, and safe automation. About the Role We’re looking for a senior cybersecurity and cyber intelligence practitioner who is also an operations leader, strategist, and people manager. You’ll bring sound cyber judgment, strong analytical and automation instincts, and the ability to stay close to the work while setting direction, managing and developing people, aligning partners, building programs from 0→1, and translating ambiguity into durable operating systems. This role combines senior individual-contributor depth with direct people-management accountability. You’ll personally shape complex cyber judgments, operating models, SOPs, quality systems, and automation strategies while building, coaching, and holding accountable a high-performing team. You’ll also lead through influence across FTEs, BPO partners, and cross-functional teams. Baseline operational health—including quality, queue coverage, productivity, and SLA adherence—is table stakes; success is additionally measured by durable improvements to decision quality, stakeholder alignment, reviewer capability, and the systems that run the operation. This role is based in San Francisco, CA. We use a hybrid work model of 3 days in the office per week and offer relocation assistance to new employees. In this role, you will: Set the strategic direction for Cyber Operations and translate high-level organizational priorities into a clear operating model, roadmaps, SOPs, escalation paths, quality goals, and safe-access strategies. Serve as the senior cybersecurity and cyber intelligence expert for complex and high-risk decisions across ChatGPT, API, Codex, agents, and

ABOUT THE TEAM Critical Harm Operations sits within User Safety & Risk Operations and builds enforcement systems for Frontier Risk and Material Harm that are accurate, fast, defensible, and built to scale. The Cyber vertical turns policy into reviewer standards, calibrated judgment, quality systems, escalation paths, and automation guardrails. ABOUT THE ROLE We are looking for a senior cybersecurity practitioner and operations strategist to raise the quality, scalability, and technical rigor of our Cyber Operations. You will combine hands-on cyber judgment with systems-level operating design: resolve the hardest dual-use questions, evolve SOPs, uplift reviewers and vendors, and build practical tools and automations. This is a senior IC role. Success is not primarily cases closed; it is durable improvement in the operating model and the reviewers who run it. IN THIS ROLE, YOU WILL: Drive the Cyber Operations operating model across domain priorities, SOPs, escalation paths, quality health, vendor capability, roadmap inputs and help inform trusted access strategies. Serve as the senior cyber expert for complex or high-risk decisions across ChatGPT, API, Codex, agents, and emerging product surfaces. Translate policy ambiguity, quality misses, appeals, and reviewer disagreement into clear decision rules, calibration examples, training, and tooling requirements. Build durable operating systems and quality loops: golden sets, holdouts, double-labeling, adjudication, error taxonomies, reviewer calibration, and automation evaluations. Raise FTE and BPO capability through onboarding, certification, coaching, recurring calibration, and vendor-performance partnership. Use quality, appeals, SLA, backlog, and disagreement signals to diagnose root causes and prioritize high-leverage fixes. Build hands-on solutions—SQL analyses, scripts, dashboards, LLM eval workflows, evidence enrichment, routing logic, and lightweight automations—that improve decision quality and reduce manua