At Netflix, our mission is to entertain the world. Together, we are writing the next episode - pushing the boundaries of storytelling, global fandom and making the unimaginable a reality. We are a dream team obsessed with the uncomfortable excitement of discovering what happens when you merge creativity, intuition and cutting-edge technology. Come be a part of what’s next. Machine Learning/Artificial Intelligence powers innovation in all areas of the business, from helping members choose through personalization, to making payment processing easier. Building highly scalable and differentiated ML infrastructure is key to accelerating this innovation. # The Opportunity Agents are one of the ways work gets done at Netflix — and the AI Platform is building the foundation they run on. The Agent Platform team owns the durable, horizontal substrate every agent at Netflix is built on, no matter which team ships it or which business it serves: the runtime agents execute in, the SDK developers build them with, the memory that lets them carry context across runs, the identity and permissioning that lets them act safely, and the tool and service interfaces they reach the rest of Netflix through. We are a small team with outsized leverage — what we ship becomes the foundation for agents across all of Netflix. The hardest problems we work on have no vendor answer and no textbook: how a non-deterministic system earns trust at member scale and how memory and evaluation become a loop that makes agents measurably better, not just bigger. This is a senior individual-contributor role for an engineer who wants to build that foundation hands-on — to own a core primitive end-to-end and set its technical direction, in a domain that is still being invented and where the frameworks turn over in months. The defining work is judgment about what to build and what actually moves the business — expressed in code and contracts the rest of the company depends on. You will: * Design, build, and operate a core agent primitive end-to-end — the runtime, the developer SDK, the memory layer, the tool and service interfaces, or the identity and permissioning substrate. You own it as a paved path other Netflix engineers ship production agents on, including the reliability and day-2 operations that non-deterministic systems running at very high — sometimes member-facing — scale demand. * Build the loop that lets agents improve from their own experience. Capture each run's full reasoning-and-tool-call trajectory as structured, replayable data; turn real-world outcomes into evaluation signal; make memory compound across runs; and close the feedback loop so improved behavior ships behind eval gates and safe rollout. These are the most durable problems in the space — make them first-class platform capabilities with real contracts, not afterthoughts. * Build the quality and safety surfaces teams can't ship agents without — the tracing, quality signals, and guardrail hooks that tell a team whether an agent is actually any good and contain non-deterministic behavior before it reaches production. * Aim the hard problems at business impact. A pull toward hard technical problems is the baseline here — the differentiator is aiming it at the ones that move Netflix's business, from personalization and content to payments and the member experience. Know where an agent actually changes a member or business outcome, and let that decide what the platform takes on and in what order. * Set technical direction and raise the bar. Make the hard architecture calls in your area, drive cross-team technical initiatives with ML scientists, product managers, and security, compute, and data partners, and define the clean interfaces a shared foundation depends on — leading through technical credibility, not authority. * Iterate fast with real users. Ship, learn from how agent developers actually build, and turn that into durable platform capabilities rather than one-off fixes. If you want the distance between "a frontier idea about how agents should work" and "the substrate every agent at Netflix runs on" to be one team, this is the role. # Minimum Qualifications * Significant software engineering experience (typically 8+ years), with a track record of shipping and operating production systems — not just prototypes. * Hands-on experience building, deploying, and operating LLM agents in production — systems that plan, call tools, observe results, and iterate — beyond chat-completion apps or demos. * Experience evaluating agents the rigorous way: building eval suites, tracing, and quality signals, then iterating on the results to make agents measurably better. * Strong fundamentals in designing and operating scalable, observable, fault-tolerant distributed systems. * A track record of building SDKs, APIs, or libraries that other engineers build on — you think in contracts, versioning, and developer experience, not just features. * Working fluency with modern agent frameworks and the tool/function-calling and MCP patterns they share — deep enough to have an opinion about where they help and where they get in the way. * Proficiency in Python and its packaging tooling, plus one of Java, Go, C/C++, Rust, or Zig. * A track record of choosing what to build by business impact, not technical interest alone — you partner closely with the teams and stakeholders you serve, and can point to outcomes that moved the business, not just systems you shipped. * Strong written and verbal communication; comfort with ambiguity; able to drive both 0-to-1 and 1-to-100 work; effective across a distributed (US) team. # Preferred Qualifications * Experience building memory or state systems for agents — context that compounds across runs — especially at high scale. * Experience with the emerging class of agent concerns: agent identity and permissioning, guardrails and safety, and cost governance for LLM and agent workloads. * Credibility in a relevant open-source ecosystem (agent frameworks, runtimes, evaluation, or developer tooling). * Familiarity with our stack — Temporal, FastAPI, PostgreSQL, Kubernetes — and with large-scale build, release, CI/CD, and observability practices. Generally, our compensation structure consists solely of an annual salary; we do not have bonuses. You choose each year how much of your compensation you want in salary versus stock options. To determine your personal top of market compensation, we rely on market indicators and consider your specific job family, background, skills, and experience to determine your compensation in the market range. The range for this role is $466,000.00 - $750,000.00. This compensation range will vary based on location. Netflix provides comprehensive benefits including Health Plans, Mental Health support, a 401(k) Retirement Plan with employer match, Stock Option Program, Disability Programs, Health Savings and Flexible Spending Accounts, Family-forming benefits, and Life and Serious Injury Benefits. We also offer paid leave of absence programs. Full-time hourly employees accrue 35 days annually for paid time off to be used for vacation, holidays, and sick paid time off. Full-time salaried employees are immediately entitled to flexible time off. See more details about our Benefits here. Netflix is a unique culture and environment. Learn more here. Inclusion is a Netflix value and we strive to host a meaningful interview experience for all candidates. If you want an accommodation/adjustment for a disability or any other reason during the hiring process, please send a request to your recruiting partner. We are an equal-opportunity employer and celebrate diversity, recognizing that diversity builds stronger teams. We approach diversity and inclusion seriously and thoughtfully. We do not discriminate on the basis of race, religion, color, ancestry, national origin, caste, sex, sexual orientation, gender, gender identity or expression, age, disability, medical condition, pregnancy, genetic makeup, marital status, or military service. Job is open for no less than 7 days and will be removed when the position is filled.