Product Manager

About Thread

No one wants to feel like just another number—especially when they need help with an urgent issue at work.

But, in a world where we can connect with family, friends, our favorite brands, and sometimes even celebrities on-demand, most service providers still ask users to drop what they’re doing, look for contact information, and choose one of two support options: bad, or worse. Either call a 1-800 number and sit on hold, or email [ help@turnitoffandonagain.com ( mailto:help@turnitoffandonagain.com ) and wait for a reply.

At best, the user’s real-time request for help turns into [Ticket: #####] in the queue; and, in an industry where service is the product, the provider signals quite clearly which customer will always matter most: “Next in line.”

At Thread, we’re on a mission to help IT service providers turn users’ worst workdays into moments of magic. We’re a VC-backed startup built by high-performance technologists with fundamental passion for service. Our heroes are people who feel most alive when doing what they’re great at in service of other people*.* We believe every thread is a call to adventure, and nothing delights us more than helping our partners deliver delightful service experiences that brighten user journeys.

The Role

Product at Thread is part of how the company decides what to build, not a ticket-writing function. The people judging our work are technicians answering calls and clearing queues all day, so every decision is felt.

We’re looking for a Product Manager to own a product area end to end, from the problem to the shipped result and the outcomes after it. You’ll report to the Head of Product & Design and work side by side with design and engineering from discovery through release.

You’ll set the direction for your area and take work from an ambiguous problem to something MSPs adopt and rely on. Much of it has no playbook yet: how an AI agent earns trust, when it acts on its own and when it asks first, how an MSP stays in control. We move fast, and we expect you to already use AI tools as a real part of how you research, write, analyze and prototype.

You won’t arrive an expert in MSPs, every tool or every phase of the work, and that’s fine. What matters is a strong appetite to learn, the humility to take hard feedback well, and the skill to give it. You’ll work with people from very different backgrounds, from support technicians to engineers to MSP owners, and get the best out of all of them.

What you’ll do

Drive agent workflows across the MSP help desk (triage, resolution, escalation and follow-up), and help decide which workflows agents take on next.

Help define the rules for agent actions: what an agent can do on its own, what needs a technician’s confirmation or approval, and how a clean handoff to a person works.

Define what “good” means for each workflow: success criteria, evals, quality thresholds and launch gates. Track resolution rate, escalation accuracy, technician time saved and customer satisfaction once it’s in production.

Shape what MSPs can configure themselves, such as policies, knowledge, escalation rules and per-client settings.

Spend time with MSP owners, service managers and technicians every week. Turn what you learn into clear priorities, and share the “why” with the team.

Support customer pilots and POC weeks, and turn the gaps you find into shipped improvements quickly.

Prototype agent behavior and flows yourself, including prompts, flows and quick builds, to test ideas before committing engineering time.

Partner with design on research and with engineering on tradeoffs between quality, latency and cost.

What we’re looking for

2 to 3 years in product management (or a mix of PM and closely related work such as founding, solutions, or technical product ops), including shipping B2B software people rely on.

Curiosity and some hands-on exposure to LLM-powered or AI agent products: personal projects, internal tools, or shipped features all count. Ideal if you’ve worked on something that takes actions, not only answers questions.

Enough practical grasp of how agent systems work (context, tool calling, evals) to read a trace, ask good questions, and make a sound call with engineers.

Good judgment about trust and control in products where mistakes have real costs.

Comfort joining and running customer research: you listen well, synthesize clearly, and change your mind when evidence says to.

Clear writing. Your docs, prompts and updates help others decide and build.

You do well in a small team where scope is broad and the path isn’t fully mapped yet.

Nice to have

Experience with MSPs, IT service management, help desk / PSA / RMM tools, or customer support operations.

Products used by service providers on behalf of their own clients (multi-tenant or partner-facing setups).

Experience with voice, chat or ticketing channels, or with MCP and agent frameworks.

You’ve shipped or deeply contributed to an AI feature in production.

How we work and what we offer

Weekly time with MSP customers. We build alongside them, not just for them.

POC weeks and beta programs to test ideas in real environments before we scale them.

Small team, high trust, close collaboration between product, design and engineering.

Based in Europe / EU time zones preferred.

Thread is an equal opportunity employer. We welcome applicants of all backgrounds and are committed to building an inclusive team.

We may use AI-enabled tools in our hiring process, such as interview transcription and note-taking. People, not AI, make every hiring decision.