AI Product Engineer

Blank Theory has spent the past year building an internal platform that sits inside how our agency operates, connecting data, automating workflows, improving reporting and giving our team AI-assisted tools they use in their day-to-day work.

The platform is live. People rely on it. Now we are hiring the person who will own what comes next. This is not a greenfield build and it is not a pure machine-learning role. You will inherit a production platform, understand how it fits into the business, work directly with the people using it, and decide what should be improved, automated or added next.

What is already built

The platform brings together agency data, workflows and AI across areas including advertising performance, servicing, reporting, billing, budget pacing, capacity planning and internal operations.

It includes production integrations, authentication and permissions, AI-assisted workflows, reporting tools and operational systems.

The environment includes React, TypeScript, Vercel, Supabase, PostgreSQL, serverless functions, model APIs, prompt management and AI observability.

You are improving an established system with real users, real data and real business impact, not starting from a blank repository.

What you will own

You will be the primary engineer and technical owner of the platform, with direct access to the people using it and meaningful influence over how it develops.

You will work with specialists across Paid Media, SEO/AIO, Content, Email Marketing, Account Management and Operations to understand how work is done, where time is lost, where errors occur and where better systems could improve the outcome.

Your job is not simply to build what someone asks for.

You will need to understand the underlying problem, decide whether it should be solved with software, automation, AI or a process change, and then design the right solution.

That could mean improving an existing workflow, building a new feature, creating an automation, introducing an AI-assisted process, strengthening an integration or deciding deterministic software is the better answer.

You will own the path from discovery to production: requirements, solution design, implementation, testing, release, monitoring and iteration.

You will also be accountable for platform reliability, data integrity, permissions, AI quality, production issues and maintainability.

Success is not measured by how many AI features you ship. It is measured by whether the agency operates better because of what you build.

Who we are looking for

You have several years of commercial software engineering experience and have shipped and supported production applications.

You are comfortable inheriting an existing product or substantial codebase, understanding how it works and improving it without needing everything rebuilt from scratch.

You can speak with non-technical users, identify the real problem behind an initial request, make sound technical trade-offs and convert incomplete requirements into a testable delivery plan.

• You have hands-on experience building LLM-powered applications and using AI coding agents as part of a disciplined development process.
• You understand prompt engineering, structured outputs, tool use, evaluation and production monitoring.
• You care about reliability. You test your work, review AI-generated code critically and investigate issues rather than treating generated output as inherently correct.

Technical capabilities

We are particularly interested in experience with:

• TypeScript, React, Supabase, Vercel, PostgreSQL and serverless architecture.
• You should understand relational database design, APIs, authentication, permissions and third-party integrations, including OAuth, rate limits, retries, scheduled jobs and data reconciliation.
• Experience with Langfuse, Anthropic, OpenAI or similar AI infrastructure is useful.
• Digital agency or marketing technology experience is helpful, but not required.

Why this opportunity is different

Blank Theory is deliberately investing in proprietary technology rather than accepting the way agencies have traditionally operated.

We want to build systems that make our people more effective, improve the quality of our work and remove unnecessary manual effort.

This is an opportunity to build software from inside the business using it.

You will not be several layers removed from the problem or waiting for a product manager to write tickets. You will work directly with department specialists and leadership, shape the roadmap and see quickly whether what you build genuinely improves how the business operates.

Application

To apply, send us your resume, examples of relevant products or technical work, and 2–3 sentences explaining how you currently use AI coding agents in your development process.

If you have inherited and improved an existing product, shipped AI functionality into production or built systems around messy real-world workflows, we would like to see how you approached it.

About Blank Theory

Blank Theory is built around people who genuinely care about their work.

We look for people who are passionate about their field, take pride in what they produce and want to keep improving. This has helped us build a talented and reliable team focused on delivering strong outcomes for our clients and supporting one another along the way.

We take the work seriously, but we do not believe the workplace needs to feel overly corporate. Our culture is relaxed, open and collaborative. People are encouraged to share ideas, ask questions and contribute beyond their job title.

You will work alongside specialists across strategy, creative, media, development, data and automation. It is an environment where good people can learn from each other, challenge how things are done and help build something better.

We offer hybrid working, regular team events and two days of volunteer leave each year. More importantly, you will be joining a team that cares about the quality of its work and is committed to building something special together.