Forward Deployed Engineer
Apollo Research
Forward Deployed Engineer
Apollo Research
Audits AI models for deceptive alignment
We are building Watcher, a coding agent security product. Watcher is deployed in production and monitors billions of agent tokens per month across engineering teams at agent-building scale-ups and multinational enterprises. Every one of those deployments involves custom engineering: integrating with the customer's coding agents and security stack, running in their cloud or on-prem environment, and tuning monitors to their workflows.
We are looking for a Forward Deployed Engineer to own that work end to end. You will be the technical face of Watcher to customers across the entire lifecycle: running technical evaluations and POCs with prospects, deploying Watcher into customer environments, building the integrations and custom capabilities they need, and providing white-glove support that turns early customers into long-term partners. You are an engineer first. When a customer needs something, your default is to build it, and then to work with the product team to decide what becomes part of the core product.
This is truly a "start-up role." You will be one of the first dedicated customer-facing engineers, you will have significant say in shaping how Apollo deploys and supports Watcher, and you can earn more responsibility quickly. As we grow, this role can develop toward leading forward-deployed engineering, solutions architecture, or core product engineering, depending on your interests.
Own customer deployments end to end: cloud, on-prem, and self-hosted backends, from architecture discussion through production rollout.
Build integrations between Watcher and customer environments: coding agent setups (Claude Code, Codex, Cursor), SIEM and security operations tooling, identity and access systems, CI/CD pipelines.
Configure and tune monitors, policies, and permissions for each customer's workflows, and validate that monitoring quality holds on their real traffic.
Debug issues in environments you don't control, often under time pressure, and drive them to resolution with the product team.
Run technical evaluations and POCs with prospects: scope the pilot, deploy Watcher on their agent traffic, and demonstrate concrete value (caught failures, security coverage, developer experience).
Serve as the technical counterpart to security and engineering leaders during evaluations: answer architecture and security questions, run demos on the prospect's actual use cases rather than canned examples.
Work with our AI Security & Control Engineer on security questionnaires, architecture reviews, and customer pen-test support during procurement.
Feed honest signal back into the team about why deals progress or stall technically.
Be the named technical contact for key customers. Own their escalations, incident communication, and feature requests.
Build customer-specific capabilities where warranted, and drive the "should this be product?" conversation with the Engineering Manager and product engineers so one-off work compounds into the roadmap.
Turn deployment experience into leverage: documentation, deployment tooling, and runbooks that make every subsequent deployment faster and less bespoke.
Represent the voice of the customer in planning: you will know better than anyone what real security and platform teams need from Watcher.
Land a lighthouse enterprise: Run a POC for a large enterprise with a strict security posture: scope the pilot with their security team, stand up a self-hosted backend inside their infrastructure, integrate with their SIEM, tune monitors on their traffic, and present results that convert the pilot into a production contract.
Build a reusable deployment path: After your first several deployments, turn the bespoke steps into a repeatable, documented, partially automated deployment package (infrastructure templates, integration modules, validation checks) that cuts time-to-production for new customers by more than half.
Ship a customer-driven capability into the product: Take a recurring customer request (e.g. a custom alert routing scheme or a new agent integration), build it for the customer who needs it, generalize it with the product team, and ship it as a core Watcher feature.
Rescue a wobbly account: A customer's monitors are noisy, their engineers are annoyed, and their security lead is questioning the renewal. Diagnose the root causes of their traffic, retune the configuration, fix the underlying issues with the product team, and win back their confidence.
Strong engineering foundation. 3+ years building and shipping production software. You can work across the stack (Python and TypeScript primarily), read unfamiliar code fast, and are comfortable with cloud infrastructure, networking basics, and deployment tooling. You will regularly build under constraints you don't choose, in environments you don't control.
Genuinely good with customers. You communicate clearly with both hands-on engineers and security leadership, you stay calm and constructive when something is on fire, and customers come away trusting you. This is not a tolerance requirement; you should like this part of the job.
High agency and ownership. When a customer is blocked, you don't wait for a ticket to be routed. You find the problem, fix it or escalate it with a proposed solution, and follow through until it's resolved.
Startup pace. You are excited about a fast-moving environment, comfortable with ambiguity and changing priorities, and willing to grind when it matters, including occasionally at customer-driven hours.
Judgment on custom vs. core. You can tell the difference between a request worth building bespoke, a request that should become a product, and a request to push back on, and you can deliver that pushback to a customer gracefully.
Strong written communication. Deployment docs, incident updates, and POC reports need to be clear and precise.
Comfortable leveraging AI heavily for engineering and customer work, while holding the output to a high standard.
Prior FDE, solutions engineering, or professional services experience, especially at a developer-tools, security, or infrastructure company.
Experience deploying software into enterprise environments: on-prem, VPC deployments, SSO/identity integration, security reviews, procurement processes.
Familiarity with the enterprise security stack (SIEM, DLP, endpoint tooling) and with how security teams evaluate vendors.
Experience with coding agents as a power user or builder. You have used Claude Code, Codex, Cursor, or similar tools heavily, or built agent tooling yourself.
Pre-sales exposure: you have run POCs or technical evaluations and know how to demonstrate value on a prospect's terms.
Willingness to travel occasionally to customer sites.
Formal AI safety background. We need excellent customer-facing engineers who can learn the AI safety context.
Sales experience or quota history. You support deals technically; you don't carry a number.
Deep expertise in every technology we use. We care about demonstrated ability to learn fast in unfamiliar environments, which is the core skill of this job.
This role offers market competitive salary, equity, and competitive benefits.
Salary: San Francisco: $222,000 - $290,000; London: £149,000 - £195,000. We will be looking to meaningfully raise salaries soon.
Our engineers effectively have an unlimited token budget. If a better result costs more compute, use it.
Flexible work hours and schedule
Unlimited vacation
Unlimited sick leave
Up to 6 months of paid parental leave
Comprehensive health, dental and vision insurance
Retirement savings with competitive employer matching (e.g. 401(k) for US employees)
Lunch, dinner, and snacks are provided for all employees on workdays
Paid work trips, including staff retreats, business trips, and relevant conferences
A yearly $1,000 (USD) professional development budget
Relocation support and visa fees (if applicable)
Time Allocation: Full-time
Location: This is an in-person role working out of our London or San Francisco office, with occasional travel to customer sites. We offer flexible working hours and some wfh arrangements.
Visa sponsorship: We sponsor visas in both the UK and US. Sponsorship isn't guaranteed for every role or candidate, but if we make you an offer, we'll work with you to find the right visa route.
The product team consists of product engineers: Jeremy Neiman, Zak Walters, Zen van Riel, Srdjan Miletic and Gustavo Bicalho; research scientists: Victor Gillioz, Monika Jotautaitė, Dmitrii Volkov; and our GTM lead: Kyle Dai. Marius Hobbhahn (CEO) advises the team. Furthermore you will interact with our other SWEs and researchers, since we intend to be "our own customer" by using our products internally for our research work. You can find our full team here.
The rapid rise in AI capabilities offers tremendous opportunities, but also presents significant risks. At Apollo Research, we're primarily concerned with risks from Loss of Control, i.e. risks coming from the model itself rather than e.g. humans misusing the AI. We're particularly concerned with deceptive alignment / scheming, a phenomenon where a model appears to be aligned but is, in fact, misaligned and capable of evading human oversight.
We work on the science of scheming, detection of scheming (e.g. building evaluations), and scheming mitigations (e.g. anti-scheming). We also work on control and monitoring research (see our scalable monitoring agenda). We work closely with many frontier AI companies, such as OpenAI, Anthropic, Google, Meta, Thinking Machines and others, e.g. to test their models and collaborate on the science of scheming. At Apollo, we aim for a culture that emphasizes truth-seeking, being goal-oriented, giving and receiving constructive feedback, and being friendly and helpful. If you're interested in more details about what it's like working at Apollo, you can find more information here.
We also build a coding agent security product called Watcher that secures agent deployments in companies. Our goal is to reduce the probability of catastrophic incidents by securing coding agents, learning about their real-world risks, and publishing our research on how to build these control systems most effectively.
Equality Statement: Apollo Research is an Equal Opportunity Employer. We value diversity and are committed to providing equal opportunities to all, regardless of age, disability, gender reassignment, marriage and civil partnership, pregnancy and maternity, race, religion or belief, sex, or sexual orientation.
Please complete the application form with your CV. The provision of a cover letter is neither required nor encouraged. Please also feel free to share links to relevant work samples.
About the interview process: Our multi-stage process includes a screening interview, a take-home test (3 hours), 3 technical interviews, and a final interview with Marius (CEO). There are no leetcode-style general coding interviews. You may use AI tools on the take-home; we judge the result the way we'd judge any contributor's work, so you are responsible for the quality of everything you submit. If you want to prepare, we suggest building simple monitors for coding agents and running them on your own Claude Code / Cursor / Codex / etc. traffic.
Your Privacy and Fairness in Our Recruitment Process: We are committed to protecting your data, ensuring fairness, and adhering to workplace fairness principles in our recruitment process. To enhance hiring efficiency, we use AI-powered tools to assist with tasks such as resume screening. These tools are designed and deployed in compliance with internationally recognized AI governance frameworks. Your personal data is handled securely and transparently. All resumes are screened by a human and final hiring decisions are made by our team. If you have questions about how your data is processed or wish to report concerns about fairness, please contact us at [email protected].
Audits AI models for deceptive alignment
Apollo Research is an AI safety organization that audits high-risk failure modes in large AI models, with a focus on deceptive alignment where models pretend to be aligned while pursuing their own objectives. It conducts fundamental research in interpretability and behavioral model evaluations, then applies these insights to audit real-world models. By combining examination of model internals with behavioral assessments, Apollo Research aims to provide stronger safety assurances than relying on behavioral evaluations alone. Its goal is to minimize catastrophic risks associated with advanced AI systems and ensure more reliable, safe behavior in deployed models.
Explore more opportunities at Apollo Research