Senior Manager of AI Practice, Claude Corps

About this role

CodePath is the largest educator of college computer science students in the country. We've trained over 40,000 students from 1,000+ universities, and our partners include Amazon, Google, and other leading technology companies. We've spent nearly a decade training the next generation of technical talent, and we just launched a $150M initiative with Anthropic, building one of the most ambitious AI workforce programs in the world.

We're now expanding into new markets and scaling our team so we can move at the speed AI is transforming the workforce. People joining CodePath now will have the opportunity to help architect the next frontier of our work.

We believe technical skill is the fastest way to turn raw talent into real opportunity. We train the engineers the AI era runs on, building toward millions of learners, hundreds of millions in revenue, and billions in economic impact. If you want to own something and be part of a 0-to-1 journey at an organization moving at the speed of AI, this is the place to build it.

About the Role

Location: Remote, United States

Role-Type: Full-Time, W2 Employee

Reports To: Senior Director, AI Programs

Compensation: $130,000 to $148,000 per year

Start Date: September 21st, ahead of the October 19th cohort launch

Together with Anthropic, CodePath launched Claude Corps in June 2026, a national fellowship program with a $150 million commitment to place 1,000 AI-trained fellows at nonprofits across America. The mentorship model works, but we need experienced professionals to coach and guide those mentors, as well as be responsible for the success of our mentor’s performance and our curriculum delivered to the Fellows. A CodePath Mentor coaches about 20 fellows, so a thousand fellows means more than forty mentors, and someone has to own whether all of them actually deliver. That someone is the Senior Manager of AI Practice.

You manage a team of mentors and you own the delivery outcome for their fellows: at full scale, about ten mentors and 200 fellows, fewer while the program ramps. You are accountable for whether fellows are actually learning, pod by pod and mentor by mentor. When a mentor's pods stall, when a fellow's work drifts off track, when the same problem shows up across three pods at once, it is yours to catch and yours to fix.

Curriculum will be developed, with your influence, at scale, you own the delivery layer: seeing where it lands across your pods and where it doesn't. You keep Commons central to how your mentors and fellows work, and you close the loop back to the curriculum team so the next version is sharper than the last.

This is a hands-on, operational role, and it grows with the program. You start by running the first cohort's small mentor team directly and build the delivery systems that let quality hold as the program climbs from 100 fellows to 1,000. A real part of the craft is reading when a lesson should come from a mentor inside a pod and when the whole group needs to hear it from you in real time. It is built for someone who has run delivery for a team before and would rather raise ten people's game than do all the work themselves.

Key Responsibilities

Manage the mentor team

•
Manage a team of mentors, about ten at full scale and fewer during the ramp: hire, onboard, coach, and hold the bar on the quality of their coaching

•
Run your team's operating cadence: weekly syncs, 1:1s, and the escalation path when a mentor or a fellow needs help

•
Make sure the day-to-day the model depends on actually happens across your team: pull requests reviewed on time, milestone reviews run, host managers engaged

Own delivery for your fellows

•
Own the delivery outcome for your portfolio, roughly 200 fellows at full scale: whether instruction lands, not whether mentors look busy

•
Own the agenda-based instructional time for your whole portfolio: set the agenda, and decide each session whether to teach it yourself, live and to everyone at once, or hand it to your mentors to run in their pods. Either way, whether it lands is on you

•
Hold the quality framework for mentorship: what good coaching looks like, how it gets reviewed, and what happens when it misses the bar

•
Own the milestone-review roll-up across your fellows, and step in on the ones whose work is at risk

•
Replace hand-tracking with delivery systems that hold at volume; the model ran on spreadsheets at 100 fellows and won't at 1,000

Own the CodePath-to-Commons layer

•
Partner with the Claude Corps strategic partners as their counterpart in the field: you see what is landing across your mentors and fellows, and you bring it back as structured signal

•
Own how your team uses Commons: the curriculum ships through the platform, and you own whether it is used the way it was designed

•
Turn what you see in the field into curriculum change requests the curriculum team can act on, and close the loop when they ship

Key Success Metrics

•
100% of your mentors are running their weekly cadence on schedule and you can pull that from your dashboard, not from what mentors report in standup

•
You can name where each mentor is strong and where they need help, and your weakest spot has a concrete plan implemented within 30 days

•
You've surfaced your top 3-5 delivery patterns across your fellows to the Claude Corps leadership team, and at least 1 has shipped as a curriculum change

•
You have a single delivery view that shows the state of your portfolio at a glance, in place of whatever was hand-tracked before you

Qualifications

Required

•
You have managed people and can point to specific ways you raised a team's performance, ideally technical or coaching staff

•
3-5 years software engineering, software engineering management, product management, or other engineering-adjacent management experience to judge the quality of technical coaching and read what fellows are shipping; you don't have to be the deepest engineer in the room, but the mentors have to respect your…