Strategic AI Lead

Snorkel AI Snorkel AI · Data AI · Redwood City, CA +1 · Remote · 410 - DaaS Delivery

This role leads high-stakes Data-as-a-Service programs end-to-end, acting as the CEO of the program. It involves owning the P&L, managing stakeholders (clients, internal teams, expert contributors), designing training data programs, developing and scaling the contributor network, building systems from scratch, monitoring performance, driving communications, managing data flows, and operating with urgency to ship better models. The role requires ML intuition for training data quality and evaluation design, and strong operational skills for managing complex programs at scale.

What you'd actually do

  1. Own the program like a CEO. Own the P&L for your pipeline: accountable for quality, timelines, revenue, margin, and budget. Forecast accurately, surface risks early, and manage spend against plan.
  2. Be the single-threaded owner across stakeholders. Your program touches client research and engineering teams, Snorkel's own forward deployed engineers, researchers, and growth partners, and the Expert Contributor Network supporting data production. You are the single person who keeps all of that aligned, accountable, and informed.
  3. Design the training data program. Translate what the model needs to learn into task definitions, quality rubrics, evaluation criteria, and RL environments. Decide what "great" looks like before expert contributors start producing it.
  4. Develop and scale the Expert Contributor Network. Define the contributor profile your program needs, work with cross-functional partners to recruit them, and design the training and QA systems that ramp them quickly. Maintain quality standards as volume grows from hundreds to thousands of contributors.
  5. Build the system from scratch. Translate vague, evolving client requirements into crisp workstreams, success criteria, and milestones. Stand up the workflows, quality systems, and operating rhythms where none exist, then harden and automate them as pipeline scales.

Skills

Required

  • Bachelor's degree in Computer Science, Information Technology, or a related field, or equivalent hands-on experience
  • 5+ years in a role where you owned outcomes end-to-end: technical program/project management, founding/early-stage operator, management consulting, or technical product management
  • A proven track record of building something from scratch in a resource-constrained, fast-changing environment, and adapting as the ground shifts
  • Analytical by default, with hands-on comfort with SQL, JSON and Python when you need to dig into data yourself.
  • ML intuition to reason about training data quality, evaluation design, and reward shaping for RL environments - and to engage as a peer with researchers at frontier labs.
  • Exceptional written and verbal communication; comfort operating with C-level sponsors and individual contributors in the same day
  • A genuine tolerance for ambiguity and a strong bias toward action

Nice to have

  • Top-tier management consulting or MBA background

What the JD emphasized

  • owned outcomes end-to-end
  • building something from scratch
  • ML intuition
  • bias toward action

Other signals

  • Data-as-a-Service programs
  • transform expert knowledge into specialized AI at scale
  • build custom AI with their data faster than ever before
  • translate technical requirement into an executable program and delivers it
  • design tasks that produce signals
  • run a multi-million-dollar program at the scale of thousands of expert contributors
  • build the system from scratch
  • harden and automate them as pipeline scales
  • Monitor, measure, and mitigate
  • Operate with urgency
  • shipping better models