Top Entry Level Machine Learning Jobs in San Francisco Bay Area, CA

Architect and ship large-scale ML and evaluation systems for Atlassian Rovo Chat. Define technical strategy for measuring helpfulness, correctness, reliability, and agentic capabilities; develop datasets, simulations, automated judges, experimentation platforms, and production quality signals. Identify failure modes, implement interventions, and measure customer impact. Lead cross-functional initiatives, establish reusable AI quality standards, and mentor engineers while balancing quality, safety, latency, reliability, and cost.

Research Engineer responsible for identifying weaknesses in voice AI models, defining improvement targets, designing evaluations, curating and generating data, fine-tuning deep learning models, and rapidly shipping production systems. The role emphasizes end-to-end ownership of customer problems, measurable production improvements, practical judgment, and fast experimentation. Candidates should have strong Python and PyTorch skills, production ML experience, and ideally experience with speech technologies, distributed training, publications, open source, or public benchmarks.

Design and build realistic simulated workplace environments, evaluation tasks, benchmarks, graders, and scoring systems for training and assessing AI models. Generate synthetic records, files, and communications; prevent models from exploiting evaluation systems; align benchmarks with real-world performance; and convert user feedback into new test cases. The role is early-career and emphasizes reinforcement learning or model post-training experience.

Optimize frontier model inference on TPUs across model code, kernels, and hardware. Responsibilities include profiling workloads, identifying bottlenecks, improving latency and throughput, developing optimized kernels, exploring quantization, building benchmarks, validating correctness, and collaborating with researchers and engineers to support new architectures and productionize performance improvements.

Develop and refine machine learning models using Python, analyze datasets with MongoDB, and integrate preprocessing and data pipelines. Evaluate models, tune hyperparameters, benchmark performance, document experiments, and deliver data-driven recommendations. Collaborate remotely with cross-functional contributors to improve and operationalize machine learning solutions in cloud or enterprise environments.

Design, develop, and implement AI/ML models; analyze large datasets; integrate AI solutions into existing systems; apply MavLink for autonomous systems; collaborate across teams and improve ML development processes.

Develop and productionize machine learning classifiers and evaluation systems for assessing model behavior. Responsibilities include supervised fine-tuning, distillation, rigorous evaluation, inference optimization, deployment, observability, reliability, and ongoing improvement. The role partners with researchers and software engineers to translate open-ended research needs into measurable ML objectives and scalable production systems, balancing model quality, latency, compute utilization, and cost.

Build and deploy production machine learning systems for quoting, material selection, planning, and factory decisions. Responsibilities include developing datasets, features, models, evaluation systems, and pipelines; monitoring quality, drift, latency, and business impact; designing safeguards and fallbacks; improving data quality and explainability; and collaborating with software engineers, operators, and domain experts.

Sets technical vision for Pinterest’s ML and media infrastructure, leading distributed training, fine-tuning, evaluation, inference, GPU optimization, content ingestion, and signal-processing architectures. Partners with AI/ML, Ads, Content, Shopping, and GenAI teams to productionize reliable, scalable platforms, drive architecture decisions and migrations, improve operational standards, and mentor senior engineers.

Develop and deploy applied AI and machine learning systems for customer-facing agents. Responsibilities include model fine-tuning, reinforcement learning, evaluation design, large-scale text and audio data work, post-training, voice AI, long-horizon optimization, and production deployment. The role translates customer and product challenges into ML problems and carries solutions from experimentation through measurable customer impact.

Develop core machine learning algorithms for unstructured data, including representation learning, clustering, outlier detection, active learning, and label-quality improvement. Apply deep learning research to customer problems, collaborate on product scalability, evolve the engineering stack, and mentor team members. The role uses Python and modern ML infrastructure across AWS, Kubernetes, FastAPI, and related technologies.

Architect and build a multi-cloud ML training platform for decentralized large-model training and inference across consumer devices and cloud instances. Responsibilities include infrastructure orchestration, distributed compute, GPU workloads, checkpointing, data streaming, network-condition simulation, node-churn management, observability, reliability, and resilient job execution. The role requires strong Python, infrastructure-as-code, Kubernetes, distributed systems, decentralized networking, and SRE experience.

Build and optimize low-level software for large machine learning workloads across specialized, multi-vendor hardware. Tune model serving for speed, throughput, reliability, and cost; write GPU kernels; design and operate heterogeneous compute clusters; monitor production systems; and work directly with customers. The role requires deep computer architecture and operating systems knowledge, infrastructure or systems experience, strong communication, on-site availability in San Francisco, and a computer science degree. Exceptional results and GPU or ML serving experience are valued.

Build and optimize end-to-end machine learning training infrastructure for large-scale multimodal robot data. Responsibilities include distributed training, data ingestion and transformation, checkpointing, orchestration, experiment tracking, sampling and curation, performance profiling, debugging data and training issues, and converting research prototypes into production infrastructure. The role may also involve robot learning, cloud and cluster infrastructure, post-training, on-robot inference optimization, CUDA or Triton kernels, and video-heavy datasets.

Architect and build terabyte-scale robotics data pipelines from fleet sensor streams through training ingestion. Curate, preprocess, annotate, version, and validate datasets; analyze data coverage, drift, redundancy, and quality; mine scenarios using embeddings and models; and develop samplers, benchmarks, and evaluation datasets. Partner with ML researchers to diagnose model failures and create targeted data mixtures and processing workflows that improve robot-learning performance.

Own the machine-learning signals and decisions governing web crawling and indexing at massive scale. Develop classifiers, rankers, and calibrated models to assess page quality, credibility, freshness, parsing success, duplicate content, and indexing value. Define ground truth for ambiguous problems, validate measurable search improvements, and collaborate across crawling, indexing, and retrieval teams to improve the quality and coverage of search results.

Build data preprocessing and generation pipelines, run model training and fine-tuning workflows, design benchmarking and evaluation suites, and operationalize research models into reproducible production workflows. The role collaborates with researchers, communicates metrics and results, and owns measurable machine learning outcomes from data preparation through delivery.

Develop mathematical models for music preferences, similarity, engagement, personalization, and discovery. Build and deploy scalable real-time recommendation and ranking systems, translate product objectives into metrics, run large-scale experiments and causal analyses, and collaborate with product and research leaders on personalization strategy.

Develop deep learning models using imitation and reinforcement learning to generate and evaluate autonomous vehicle trajectories. Build production machine learning pipelines, metrics, and error-analysis tools, leveraging large-scale infrastructure. Collaborate with Planning, Perception, Simulation, and Systems Engineering teams to improve autonomous driving behavior in complex urban environments.

Build and deploy scalable AI/ML systems for retrieval, ranking, categorization, triage, and generative AI across healthcare data. Own solutions end-to-end from problem framing and model selection through production deployment and optimization. Collaborate with software engineers, product managers, clinicians, and operations teams to translate complex business problems into explainable, maintainable systems with measurable outcomes.

Build and deploy production machine learning systems, including LLMs, agents, and RAG applications, for healthcare, government, energy, and other regulated industries. Responsibilities include developing scalable data pipelines, optimizing model performance, monitoring deployed models, translating research into practical solutions, collaborating with leaders and cross-functional teams, and contributing to software engineering practices. The role is designed for a 2027 bachelor's graduate with hands-on generative AI and machine learning framework experience.

Build and own Clera’s AI/ML intelligence layer, including recommendation systems, ranking algorithms, and LLM-powered candidate-role matching. Develop production-ready features, data pipelines, and services while collaborating cross-functionally and shipping frequently. The role requires hands-on experience with the listed development tools, TypeScript, React, startup environments, and a willingness to work six days per week in a San Francisco hacker house.

Own the infrastructure powering language-model training and serving from experimentation through production. Optimize GPU performance, latency, throughput, memory, and cost; improve batching, caching, compression, and decoding; support inference for evaluation, synthetic data, and reinforcement learning; and build reliable model release and observability workflows. Partner with founders and researchers to allocate GPU capacity and deploy models in customer environments.

Build and scale pipelines that transform conversation data into structured customer memory. Develop LLM extraction and retrieval systems using embeddings, vector search, and ranking; create evaluation benchmarks and regression gates; diagnose production issues through logs, metrics, and traces; and partner with Product and ML teams to improve customer-focused AI systems.

Develop, train, evaluate, debug, and improve machine learning models for real-world applications. Own the ML lifecycle from dataset construction and experimentation through deployment, while applying deep learning, reinforcement learning, and LLM techniques. Collaborate with engineering and product teams, contribute to scalable ML infrastructure and tooling, define evaluation metrics, analyze errors, and translate research concepts into practical solutions.