AI Jobs in San Francisco

View detail

Build end-to-end product experiences that span frontend applications, backend services, agent workflows, cloud infrastructure, and developer tooling. Design AI-powered workflows that generalize across a wide variety of software engineering teams, languages, codebases, and development practices. Discover and implement novel ways to apply AI to eliminate friction throughout the software development lifecycle. Partner closely with product, design, and research to understand developer needs and rapidly translate insights into shipped product improvements. Work directly with users—including developers at OpenAI, open-source contributors, startups, and large enterprises—to understand pain points and validate solutions. Improve the reliability, observability, scalability, and performance of the systems and workflows you build.

View detail

Build core primitives end to end including entity ownership, audit, authorization, and orchestration, ensuring the right actions are the default and incorrect actions are difficult. Own the domain model by turning Fluidstack's concepts of power, datacenters, and chips into composable entities that remain durable over time. Define interactions with external systems by interfacing with vendor systems and ingesting domain-specific formats such as KMZ, BIM, Revit, and vendor documents. Enable AI agents to be first-class operators of Fluidstack's systems by providing tools, guardrails, and audit trails to allow safe and effective operation beyond mere advice.

View detail

Develop AI-powered vehicle intelligence features that understand user intent, trip goals, vehicle state, and system constraints. Apply reinforcement learning, planning, optimization, and data-driven modeling to improve vehicle-level decisions across energy, comfort, charging, routing, and proactive vehicle preparation. Build models using vehicle telemetry, navigation data, user behavior, weather, traffic, cabin conditions, charging patterns, and fleet data. Create personalization models that learn user routines, comfort preferences, driving patterns, charging habits, and trip priorities while preserving privacy and user control. Use simulation, digital twins, and scenario-based testing to train, evaluate, and validate AI behavior before production deployment. Collaborate with autonomous driving and VLA teams to define interfaces for sharing user intent, route objectives, vehicle constraints, energy targets, comfort preferences, and system-level recommendations. Integrate ML models into production vehicle and cloud platforms, considering latency, compute efficiency, reliability, safety, explainability, and over-the-air update readiness. Work cross-functionally with Product, UX, Systems Engineering and Controls.

View detail

As the Applied Data Science & Insights Lead for GTM Intelligence Solutions and Technical Success, you will be responsible for shaping how OpenAI measures, understands, and improves customer adoption across B2B products by building AI/ML-powered intelligence products that integrate various customer and product data into practical operating systems for GTM and Technical Success. You will define and lead the roadmap for GTM Intelligence and Technical Success insight products, build the data science foundation including metrics and models, develop propensity score models, and create predictive and causal models related to customer health, expansion propensity, churn risk, and intervention effectiveness. You will design next-best-action systems, partner with Technical Success leaders to enumerate playbooks and measure outcomes, develop customer segmentation and benchmarking frameworks, and create scalable insight products embedded into field workflows. Additionally, you will build and lead a small team of data scientists and analytics partners, set technical standards, create team operating rhythms, maintain analytical rigor, and collaborate with multiple departments such as Data Engineering and RevOps to improve data foundations.

View detail

Translate cutting-edge research into production-ready machine learning systems. Design, build, and deploy end-to-end ML models and pipelines. Develop and optimize models for image and video processing. Own the full ML lifecycle including experimentation, training/fine-tuning, evaluation, and deployment. Rapidly prototype using open-source models and adapt them for product needs. Conduct experiments, analyze results, and iterate to improve performance. Collaborate with researchers and cross-functional teams (product, engineering, design) to deliver ML solutions at scale. Participate with advancements in machine learning and apply them to continuously improve products.

View detail

Utilize proprietary software to provide accurate input and labels for healthcare and administration projects, ensuring high-quality data for AI model training. Deliver curated, high-quality data for scenarios involving patient care coordination, medical billing, administrative workflows, and healthcare operations. Collaborate with technical staff to support the training of new AI tasks and contribute to the development of innovative technologies. Assist in designing and improving efficient annotation tools tailored for healthcare and administration data. Select and analyze complex problems in healthcare and administration fields aligned with your expertise to enhance AI model performance. Interpret, analyze, and execute tasks based on evolving instructions, maintaining precision and adaptability.

View detail

Architect and build next-generation system protections through hands-on design, model training, and deployment strategies. Lead and manage a small, senior team of Engineers, providing clear direction and autonomy. Collaborate with Research, Safety, Product, and Policy teams to use existing tools and advance new solutions. Utilize state-of-the-art models to detect and prevent problematic content. Establish evaluation frameworks and metrics to measure progress and identify improvement areas. Support team growth and maintain high performance through mentorship and career guidance.

View detail

Build backend and infrastructure systems that extend OpenAI's API platform into cloud-native environments such as AWS. Design and ship cloud-contained products that allow customers to use OpenAI capabilities while keeping workloads and data within cloud environments. Help stand up cloud-hosted Codex experiences powered by the OpenAI Responses API. Build infrastructure and runtime abstractions for a stateful, cloud-optimized agentic platform. Partner closely with external cloud partners as well as internal teams across Codex, Research, and Safety Systems to translate emerging capabilities into production-ready systems. Improve the reliability, scalability, observability, and operational maturity of the services underpinning these products. Help shape the technical direction of a new and growing team as it scales from an early core group into a larger engineering organization. This role also involves building backend services, APIs, SDK integrations, authentication flows, and cloud service infrastructure that let developers use OpenAI capabilities in the cloud environments where they already build, and working across teams sometimes embedded with partner product groups to ship products quickly across multiple platforms at the same time.

View detail

As a member of the Agent Post-training Personality team, the role involves helping to make OpenAI’s agents exceptional collaborators by studying what makes an agent thoughtful, clear, perceptive, appropriately proactive, and easy to work with. This includes translating those insights into evaluations, training data, reward signals, and model improvements. Responsibilities include developing a rigorous understanding of effective agent collaboration across various types of work, turning qualitative judgments about model behavior into concrete hypotheses, evaluations, graders, and training interventions, studying user signals to understand behaviors that create trust and satisfaction, working with human experts and trainers to produce high-quality data capturing excellent collaborative behavior, improving reward models and reinforcement learning objectives, collaborating with pretraining and early-training teams on data and objectives, building pipelines for updating training data, partnering with product teams to turn consumer insights into model improvements, and owning projects end to end from identifying behavioral failures through experimentation, training, evaluation, and launch.

View detail

Translate business requirements into AI/ML model requirements. Prepare data to train and evaluate AI/ML/DL models. Build AI/ML/DL models using state-of-the-art algorithms, especially transformers, sometimes leveraging existing algorithms from research. Test and evaluate models, benchmark quality, and publish models, datasets, and evaluations. Deploy models in production by containerizing them. Work with customers and internal employees to refine model quality. Establish continuous learning pipelines for models with online or transfer learning. Build and deploy containerized applications on cloud or on-premise environments.