Founding Machine Learning Engineer

Acerca de

the RoleThis is a founding-level, hands-on ML engineering role at a fast-growing Series A AI startup in the Bay Area, specializing in high-quality training data, post-training data pipelines, and model evaluation for frontier AI labs and enterprises

You'll work closely with leadership to design, train, and ship production-grade ML systems — while helping shape the technical culture and infrastructure from the ground up.

This role is fully on-site in Mountain View, CA.

Candidates must be authorized to work in the US without visa sponsorship.

What You'll DoBuild and optimize end-to-end ML pipelines, from data ingestion through to production deployment.

Implement and fine-tune LLMs, embeddings, and generative models for real-world applications.

• Develop efficient training and inference systems leveraging distributed compute at scale.

Partner with data and product teams to translate ideas into measurable ML impact.

Contribute to model monitoring, evaluation, and continual learning frameworks.

Establish best practices in model versioning, reproducibility, and scalability across the organization.

What We're Looking ForRequired3–10 years of hands-on experience as an ML Engineer, Applied Scientist, or Research Engineer.

Proficiency in Python and at least one of PyTorch, TensorFlow, or JAX.

Proven experience building production-grade, end-to-end ML pipelines (data ingestion, training, deployment).

Experience implementing and fine-tuning LLMs, embeddings, and generative models for real-world use cases.

Hands-on experience with distributed training/inference and scalable ML systems.

Familiarity with cloud platforms (AWS, GCP, or Azure) and ML tooling such as MLflow or Weights & Biases.

Strong collaboration.

Habilidades

with cross-functional data and product teams.Bias for action, ability to work autonomously, and eagerness to build from scratch.Nice to HaveExperience with RL environments or evaluation frameworks for AI models.Background in agentic systems, tool-use pipelines, or multi-st...