Founding Engineer - Software (Data/ML Infra)

 

Founding Engineer - Software (Data/ML Infra)

Location: San Mateo, CA
Company Stage of Funding: Seed Stage / Robotics + Embodied AI Startup
Office Type: On-site (5 days/week in San Mateo, CA)
Salary: $180,000 – $220,000 Base
Equity: 0.75% – 1.25%
Bonus: Competitive total compensation package
Visa: H1B Transfers, TN, OPT, and O-1 supported (No new H1B sponsorship)

---

Company Description

Our client is building the next generation of embodied AI systems for robotic manipulation in the apparel industry.

The company is developing intelligent robots capable of solving one of the hardest challenges in robotics: dexterous handling of deformable materials in real-world environments.

Founded in 2025 by experienced AI and computer vision leaders behind Atlas Wearables (acquired by Peloton) and Peloton’s AI organization.

Raised approximately $8.5M from top-tier venture capital firms and the National Science Foundation.

Already piloting robotic systems with customers and preparing for broader deployments across apparel manufacturing facilities.

The company is focused on applying physical AI to a massive industry with over $400B in annual apparel manufacturing spend.

The platform combines:

• Robotics data infrastructure
• Multimodal data pipelines
• ML training systems
• Distributed training infrastructure
• Computer vision
• Embodied AI
• Experiment tracking systems
• GPU orchestration platforms

Engineering challenges involve building scalable multimodal data pipelines, distributed training infrastructure, cloud data systems, GPU utilization optimization, experiment management tooling, and ML operations platforms powering frontier robotics models.

The company operates at the intersection of robotics, computer vision, machine learning infrastructure, and real-world deployment.

The culture values ownership, technical excellence, startup execution, rapid iteration, and solving practical embodied AI problems in production environments.

---

What You Will Do

• Build and maintain large-scale multimodal data pipelines supporting robotics model development
• Design systems for ingesting, processing, organizing, and querying robotics datasets
• Build infrastructure supporting teleoperation, model training, and deployment workflows
• Manage data versioning, lineage, and reproducibility across model training runs
• Develop internal tooling for exploration and analysis of robotics field logs
• Build and optimize ML training workflows including scheduling, resource allocation, and experiment tracking
• Design cloud infrastructure supporting model training at scale
• Optimize data loading and I/O performance to maximize GPU utilization
• Build distributed training infrastructure as model complexity and dataset sizes grow
• Design compute orchestration systems for large-scale training workloads
• Manage cloud storage architectures for multimodal robotics datasets
• Improve data streaming and processing infrastructure
• Profile system bottlenecks and optimize training throughput
• Collaborate closely with founders on technical roadmap and infrastructure strategy
• Support deployments and occasional customer site visits
• Own infrastructure systems from architecture through production deployment
• Help establish engineering best practices and technical foundations as the company scales
• Enable rapid iteration on embodied AI model development

---

Ideal Candidate Background

• 2+ years of experience building data infrastructure or ML infrastructure for model training systems
• 4–5+ years of overall software engineering experience preferred
• Experience building large-scale multimodal data pipelines
• Experience working with robotics, autonomous vehicles, drones, or similar physical AI domains
• Experience building infrastructure supporting ML model training workflows
• Strong software engineering fundamentals and systems design skills
• Experience building systems from 0-to-1 in startup or fast-paced environments
• Experience with cloud storage architectures and compute orchestration
• Experience optimizing I/O throughput and GPU utilization
• Familiarity with distributed training infrastructure
• Experience working with large-scale video, image, sensor, or multimodal datasets
• Strong Python engineering experience
• Experience operating production infrastructure at scale
• Ability to independently own systems end-to-end
• BS in Computer Science or related technical field
• Comfortable working onsite in a highly collaborative startup environment

---

Strong Signals

• Experience at companies such as Waymo, Cruise, Aurora, Nuro, Zoox, Covariant, Skild AI, Physical Intelligence, NVIDIA, Google DeepMind, Meta, or similar organizations
• Experience building multimodal data pipelines for robotics, autonomous vehicles, drones, or embodied AI systems
• Experience with distributed training frameworks such as DeepSpeed, FSDP, Triton, or similar technologies
• Strong PyTorch experience
• Experience with Spark, gRPC, and large-scale data processing systems
• Experience managing GPU clusters and training infrastructure
• Experience with experiment tracking platforms such as W&B or MLflow
• Experience with Kubernetes, Docker, and cloud-native infrastructure
• Familiarity with robotics datasets, sensor data, and computer vision workflows
• Experience optimizing training performance and infrastructure efficiency
• Startup experience with significant ownership and 0-to-1 execution
• Experience building ML infrastructure from scratch
• Strong interest in robotics and embodied AI
• Proven ability to operate independently in high-growth startup environments
• Experience supporting production deployments of AI systems

---

Compensation and Benefits

• Base salary: $180,000 – $220,000
• Equity: 0.75% – 1.25%
• Opportunity to join a well-funded embodied AI startup
• Significant ownership over data and ML infrastructure architecture
• Direct collaboration with experienced repeat founders
• Exposure to cutting-edge robotics, computer vision, and embodied AI systems
• High autonomy and end-to-end ownership
• Opportunity to define core infrastructure powering robotic intelligence
• Work alongside highly technical AI and robotics engineers
• Onsite work environment in San Mateo, CA
• Opportunity to help solve one of the hardest problems in robotics
• Exposure to real-world deployments and customer-facing robotics applications
• Fast-moving startup culture with substantial technical scope

---

Why Join

This is an opportunity to build the foundational infrastructure powering the next generation of embodied AI systems.

Rather than working on traditional software products, the company is solving one of the hardest challenges in robotics: enabling intelligent machines to understand and manipulate deformable objects in real-world environments.

You'll work alongside accomplished founders with deep expertise in computer vision, machine learning, and robotics while helping define the infrastructure that enables frontier AI models to learn from multimodal robotic data.

The role offers a rare combination of ML infrastructure, distributed systems, robotics, multimodal data engineering, and startup ownership.

If you enjoy building large-scale data systems, optimizing training infrastructure, working on robotics and embodied AI, and operating in high-ownership startup environments, this role offers exceptional scope and career upside.