Senior Technical Program Manager, Applied AI [NYC or SF]

About the role

Our Applied AI team is a group of forward deployed engineers (FDEs). They take Concurrence from signed contract to production at our customers, usually across several accounts at once. Each account has its own data, integrations, clinical workflows, and deadlines. This role reports to the Head of Applied AI. Our FDEs are excellent builders. What we need now is someone who makes the whole operation legible. That means knowing at any moment what every FDE is working on, what's at risk, what's blocked and why, and whether we'll hit what we've committed to. This is a hygiene role in the best sense. If work isn't tracked in Linear, we can't plan capacity, judge workloads, measure on-time delivery, or help people get better at estimating and meeting dates. You'll make Linear the trustworthy source of truth for applied work. You'll do it with the follow-through, judgment, and technical credibility that make engineers see it as useful rather than bureaucratic. You've been an engineer, and that matters. When an FDE says they're "blocked on platform," you need to know whether that's true. Is it a real platform gap, or applied-side work that hasn't been done? You need to be able to push back credibly in either direction. Above all, you have relentless follow-through. You don't let things drop, you don't wait to be asked, and you don't stop chasing until it's done.

Responsibilities

• The cross-account delivery plan. You'll own the plan across every active deployment: dates, dependencies, FDE capacity, and risk. You're the person who sees conflicts coming two weeks before they hit.
• Linear as the source of truth. You'll design how Applied AI works in Linear and then hold the team to it. That includes the project structure per account, labels that separate applied blockers from platform blockers, an estimation scale, and a shared definition of done. The result: tickets are current, statuses are honest, and due dates are set and met.
• Accountability on progress. You'll check in on work before it becomes a surprise. You'll chase updates, surface slips early, and make sure nothing goes quiet.
• Capacity and workload. You'll know how loaded every FDE really is. That includes the unplanned work that never shows up on a roadmap, like production issues, customer escalations, and safety reviews on live agents. You'll build plans with buffer grounded in real data.
• Fair delivery metrics. You'll measure on-time delivery, estimation accuracy, and how early risks get flagged. When something slips, you'll attribute it honestly to customer, platform, or applied causes, so people are judged on what they control.
• The field-to-platform loop. You'll triage blockers between applied and platform, and turn field needs into well-specified