Head of Platform Engineering

The Role

Reducto's platform is what turns a customer's document into a result. Underneath it is advanced job orchestration, document pre and post processing, reliable and fast model inference, along with the deployment platforms that let a bank or an insurer run all of it inside their own environment. We process a billion pages a month, and the customers driving that volume run us in production on workflows where a failure is their incident as much as ours. We're hiring a Head of Platform Engineering that will own the reliability of the platform, shape the future direction, and build the team that executes it.

This is a hands-on role. You'll be in design reviews, on incidents, and in the code when it unblocks the team, and you'll spend your first few months shipping as an engineer alongside the team before you fully take on leading it. As AI makes it easier to write code, the expectation that you can solve the hard problems yourself becomes more of the norm.

You'll report to our VP of Engineering and work directly with our founders.

We would love to meet you if you:

• Operating record: You have at least 3 years of experience owning reliability, latency, and efficiency for a platform of significant scale, and you can talk about the numbers you moved and how.
• Hands-on: You've been in hands-on roles recently, and you're comfortable writing design docs, writing code, and reviewing your team's code with a critical eye.
• Hiring: You have a track record of attracting, hiring, and retaining top tier talent, and you can point to people you brought in who are still doing their best work.
• Ambiguity: You can find your bearings in environments that move extremely fast and where not everything is spelled out yet. Generally this looks like having done 0 to 1 as a founder, a founding engineer, or the person who started a greenfield project and carried it through.
• Ownership: You expect to own the roadmap and the technical decision making for your team, and you'd rather bring both to planning than be handed them.
• Stack: You're familiar with infrastructure-as-code and containerized deployments, and you've operated them in production.
• Bar: You have a high bar for how things are done technically, you hold your team to it, and you apply it to your own work first.
• Curiosity: You're deeply curious and eager to learn new technology, stacks, and products, including the ones we haven't picked yet.

The Core Work Will Include:

• Owning Reducto's serving platform end to end, from job orchestration and document pre/post processing through model inference, autoscaling, and throttling, including finishing an in-flight migration of the serving stack
• Building out how we deploy and serve Reducto to customer clouds and on-premise environments
• Building the observability that tells us and our customers what the platform is doing, and using it to drive alerting, incident response, and the reliability, latency, and cost-per-document numbers the team is measured on
• Owning the platform roadmap and the technical decisions behind it, and sourcing and recruiting the platform engineers who will execute it
• Working with our Product engineering team to build and scale our agent runtimes and environments
• Building our internal data platform that serves internal users and customers
• Working closely with our ML and Product engineering teams to roll out new models and deployment capabilities

Bonus points if you

• Have served ML models, particularly vision-language models, at scale
• Have built data platforms for internal users and ML teams
• Have built, shipped, and managed on-premise deployments
• Have experience with agent sandboxes, harnesses, and evaluations

This is an in person role at our office in SF. We're an early stage company which means that the role requires working hard and moving quickly. Please only apply if that excites you.