Senior Engineering Manager, ML Platform

We're looking for a Senior Engineering Manager to lead our ML Platform Team - a growing team responsible for the foundational infrastructure that powers our machine learning work. This is a player-coach role: you'll set technical direction and contribute hands-on while building out the team and establishing the processes that will scale with it.

The platform is in its early stages, with some foundations in place. You'll be joining at a pivotal moment - making architectural decisions that will shape how the team and the platform grow from 4 engineers today to a team of 10–12.

What You'll Work On

Infrastructure Leadership

• Own the strategy, roadmap, and execution for GPU compute infrastructure, ensuring it scales to meet growing model training and fine-tuning demands

• Contribute directly to infrastructure design and implementation, particularly in the near term as the team grows

• Drive reliability, performance, and cost efficiency across distributed training clusters.  Optimize existing and new training workloads to achieve scale.

• Evaluate and adopt new hardware (GPUs, TPUs, custom accelerators) and cloud/on-prem infrastructure as the team's needs evolve

Data Platform Ownership

• Oversee the design and operation of data storage, indexing, and retrieval systems that support large-scale dataset generation

• Ensure data pipelines are performant, fault-tolerant, and meet the quality and freshness requirements of ML teams

• Establish early-stage standards for data access, lineage, and governance — pragmatic and scalable, not over-engineered

Shared Tooling & Developer Experience

• Lead the development and maintenance of shared libraries and frameworks for data transformation pipelines

• Partner with ML researchers and engineers to understand their workflows and translate them into reliable, reusable platform capabilities

• Champion developer productivity - reduce friction for teams consuming platform services

Technical Strategy & Architecture

• Lay the architectural foundations of the platform, making decisions that are pragmatic today but designed to scale to a 10–12 person team and beyond

• Make key architectural decisions around compute orchestration (e.g. Kubernetes, Slurm, Ray), storage systems, and pipeline frameworks

• Balance short-term delivery with long-term platform health -knowing when to build, buy, or borrow

Cross-functional Collaboration

• Act as a technical partner to ML research, data engineering, and product teams - translating needs into platform priorities

• Communicate roadmap, incidents, and technical tradeoffs clearly to both engineers and senior leadership

• Help ML teams become self-sufficient on the platform, reducing bottlenecks on the platform team itself

Team Building & Management

• Actively participate in hiring to grow the team from 4 to ~10–12 engineers, including defining roles and leveling

• Mentor and develop engineers, establishing a team culture early that will hold as headcount scales

• Define lightweight but durable team processes - on-call rotations, incident response, and engineering standards that won't need to be rebuilt at scale

• Be comfortable doing IC work yourself while simultaneously building the team's capacity to take it on

What We're Looking For

• 7–12 years of engineering experience, with at least 2–3 years in a formal management or tech lead capacity

• Demonstrated experience building or scaling a platform, infrastructure, or ML systems team from the ground up

• Technical credibility in one or more of: GPU/distributed compute infrastructure, large-scale data storage and retrieval, or data pipeline frameworks

• Experience making foundational architectural decisions in an early-stage or greenfield environment

• Strong cross-functional communication skills - able to translate between ML researchers, engineers, and senior leadership

• Comfortable with ambiguity; able to define the roadmap rather than just execute against one

• A hands-on mindset - willing and able to write code, review designs, and debug production issues alon