Delivery Lead

You are a senior, AI-native delivery lead who owns engagement outcomes end to end: shaping the solution, defining and planning the work, leading the client, and making sure what we build is adopted and delivers measurable value.

Why this role exists

We're building a small, senior delivery team that delivers what much larger teams traditionally would, by using AI across the full surface of the work.

AI has made building faster. It hasn't made it any easier to know what to build, get the right decisions made, or get people to adopt what's been built. That's where engagements succeed or fail, and it's what this role owns: making sure the right thing gets built, it lands, and the client's team comes out the other side working differently.

You partner with a Technical Lead / AI-Forward Engineer, working as one team: you bring the problem, the outcome, the sequencing, and the adoption; the TL brings the architecture and the build; both of you own whether it works.

This is a senior individual contributor role: you lead engagement teams, not a reporting line. Teams are small (typically 2-4 people), you run one or two engagements at a time, and clients range across mid-market industries. Engagements run from weeks-long plan-and-define and POC work to multi-year managed services. Fastloop's own delivery is the first proving ground: we are becoming AI-native at the same time as the clients we serve.

What you'll own

• Taste and judgment. Plausible plans, backlogs, and status reports are now cheap to produce. Knowing whether a story is ready, a plan is real, "on track" is true, and the work solves the problem is not. Your judgment on what's ready, done, and at risk is the bar.
• Solution shaping. You lead discovery with the TL and shape the solution together: the problem worth solving, the outcome worth paying for, what comes first, and what the client can absorb. You challenge the ask when there's a better way, including when that way isn't AI.
• Workflow redesign. You map how work actually flows across people and systems and redesign it: where agents act, where deterministic automation fits, and where a human decides.
• Documentation and breakdown. You own how work is specified: requirements, stories, and acceptance criteria clear enough for an agent or engineer to execute without a meeting, sliced vertically, with testable definitions of done.
• Planning and visibility. You plan for AI-speed delivery, where the constraints are review, validation, decision latency, and adoption rather than build time. Status comes from the source of truth, not a hand-built deck, and you re-plan from evidence.
• Team transformation. You change how client teams document, break down, plan, and report work, connecting AI into their tracker, source control, and CI/CD, and growing adoption beyond the pilot team.
• Quality and governance. You define ready and done, what "good" means for AI outputs in business terms, where humans must approve, and how sensitive data is handled.
• Outcomes and value. You agree the business outcome up front (cost, cycle time, quality, throughput, adoption), measure it, and report against it.
• Engagement and account. You own scope, risk, dependencies, resourcing, margin, and delivery health. You lead client conversations from working sessions to steering committees, grow accounts through the quality of delivery and a roadmap clients want to fund, and support pre-sales with discovery, estimates, and SOWs.
• Delivery library. You grow the team's shared library of delivery skills, agents, templates, and playbooks, and with the engagement sponsor turn proven patterns into Fastloop offerings.

How you work: the compounding loop

This is how we expect the work above to get done. You don't do it by hand: you build and run the machine that does it, own the judgment on what comes out, and contribute back so it gets better. That's the difference between an orchestrator and an operator who uses AI.

Specs, breakdowns, plans, forecasts, and status are produced by skills and agents you build and run. Your time goes to the judgment calls, the client, and making the machine better. The engagement you land matters; so does what Fastloop and the client can do next time that they couldn't before.

• Start from golden paths. Begin every engagement from the team's existing skills and playbooks (story development, scoping, status, discovery) and improve them. Rebuilding a template the team already has creates drag for everyone who comes after you.
• Contribute back continuously. A better story format, a sharper estimation heuristic, a status agent