Cogent AI Fellowship

About Cogent

Cogent is an Applied AI Lab building the next generation of AI agents for cybersecurity. AI has fundamentally changed how attacks happen, allowing malicious actors to operate at unprecedented speed and scale. Cogent’s "AI Taskforce" assesses petabytes of enterprise data to remediate these issues before critical breaches occur.

To stay at the cutting edge, we blend frontier research with real-world execution. Alongside our core product work, Cogent Research serves as our applied AI lab, providing the research horsepower needed to make truly agentic security workflows a reality.

Since coming out of stealth, Cogent has experienced rapid growth. We partner with Fortune 500 companies to secure some of the most complex production environments in the world.

We’re backed by Greylock and we’ve built a team with the best minds in applied AI. Our team is comprised of people from:

• Top universities like Stanford, Berkeley, Penn, Duke, Carnegie Mellon, Waterloo

• Preeminent research labs like Deepmind and SAIL

• Unicorn, high-growth companies like Scale AI, Databricks, Stripe, Tesla, Coinbase

• World class cybersecurity experts from Wiz, Abnormal AI, Zscaler

About the Role

Cogent Research is investing equally in two core research areas. The first is agentic security: as AI agents are deployed in production, real-world incidents have shown that prompt-level guardrails are insufficient. Agent behavior requires guarantees that are non-bypassable, AI-synthesized, and human-auditable. The second is auto-remediation: resolving security findings at machine speed in real customer environments.

Each area is a first-class research objective, and the two are deeply connected. Auto-remediation can only operate safely at the required level of autonomy when agentic security guarantees are in place. Across both areas, we combine AI with formal methods: tool calls governed by runtime-enforced policies, and code fixes that ship with machine-checkable certificates of correctness.

The Cogent AI Fellowship is an early-career research-engineering track for recently graduated, or soon-to-graduate, PhDs who want to move research from paper to production. Fellows choose a problem they are excited to pursue, prototype quickly alongside our research and engineering teams, and ship production systems. We look for people who are comfortable with ambiguity and willing to change direction when the evidence calls for it.

What You’ll Do

• Define and pursue a research direction at the intersection of AI and formal methods, such as autoformalizing intent into machine-checkable specifications, certificate-producing code generation, AI-assisted proof in Lean, runtime verification of agent tool calls, or natural-language-to-policy synthesis.

• Take ideas end-to-end: from prototype to systems running against real customer workloads.

• Partner with applied AI, agent, and platform engineers to translate research advances into production capabilities.

• Work with design partners and security leaders to ground the work in the workflows security engineers actually run.

• Build evaluation harnesses and benchmarks that capture what “verified-correct” or “policy-compliant” means in messy, real-world security environments.

• Iterate quickly, update your approach when results indicate a better path, and share what you learn internally and at venues that advance the field.

About You

• Research experience in one or more of the following areas: formal methods, program analysis, programming languages, agentic AI for code reasoning, or runtime verification.

• A strong publication record in top conferences or journals, or comparable open-source or artifact contributions.

• A recently completed PhD in Computer Science, or a closely related field, or a defense expected soon.

• Strong engineering ability, including fluency in a major programming language and the judgment to mature a prototype into a production-ready system.

• An independent, self-starting mindset; comfort operating in ambiguous settings; and genuine excitement for Cogent’s mission.

Bonus Points

• Experience with vulnerability detection/remediation or secure-by-construction systems.

For California Based Applicants

The standard base salary range for this position is $100,000 - $300,000 annually. Compensation offered will be determined by factors such as location, job level, job-related knowledge, skills, and experience. Certain roles may be eligible for variable compensation, equity, and benefits.

We are committed to building an inclusive and diverse company. We do not discriminate based on gender, ethnicity, sexual orientation, religion, civil or family status, age, disability, or race.