The role PortSwigger is going all-in on AI. Not as a side project, not as a slogan - as the central way we intend to work across every part of the business. This role is one of a small number of people who will help us actually do that. You’ll drop into hard, unscoped AI problems across the business, usually in the parts that haven’t figured AI out yet. You’ll work alongside the people closest to the work, find the bit worth solving and ship something that makes things better. In a given stretch, you might design an agentic workflow for enterprise sales, coach a leader on how their team should use AI, talk MCP connector risk with a security engineer, or rethink leadership in an AI-native organisation. Some problems take days; others grow into much bigger transformations. You’ll be scoping, designing and building with the domain experts, breaking things, testing things, laughing and swearing through the 99 failures before the run that finally works. Roving, not tribe-aligned. Always in the work, never stuck only writing slides about it. About you You’re fanatical about AI. It’s rewired how you live. You wake up and check what dropped overnight. Your browser history is release notes and changelogs and half a dozen Discord tabs. If a new model or tool landed, half your morning’s gone to testing whether it beats what you’re already using. The tool you wanted didn’t exist, so you built it over the weekend. Your home is full of things you’ve stitched together - the budget tracker, the meal planner, the story generator for your kid - but that was the starting point. Your tinkering has levelled up since: agents that run while you sleep, systems plugged into each other, tools with real users who depend on them. You have personal subscriptions to several AI products and you’ve played with many more. You switch tools when the industry moves - you’re not loyal to anything that stops being the best. Your opinions on frontier tools and models goes deep. You can explain where each earns its place to technical and non-technical people alike. You’re working smarter than you ever have - and loving it. You can’t shut up about it. Friends gently take the piss. You don’t mind, because you can see what’s coming and they can’t yet. You’re probably a bit stuck, though. Your current employer treats AI as a side project - blocked tools, capped spend, missing ambition. You want to be somewhere that takes it seriously. That’s the bit that’s going to make this role grab you. And you don’t just talk about AI - you build for it. You take ambiguous briefs, find the parts worth solving and produce something under your own steam. You don’t quite trust your own demo, so you measured whether it helped - because you’re as interested in the evals as the build. You leave an artefact trail - repos, prototypes, write-ups, side projects - and that’s how we’ll get to know you: we read what you’ve made, not just what you say about it. You can’t help seeing the workflow behind the task. Put a shiny AI demo in front of you and your first question isn’t “what could I build with this?” - it’s “what should this whole process look like now?” Basic prompting is old ground. You’re more interested in whether something genuinely works for someone than whether it demos well. The nerdy bits fascinate you - the knowledge and context behind the system, how the agents work together and whether anyone is checking that their output is actually any good. You find yourself thinking about where human judgement belongs and what all this means for teams, culture and leadership. You have strong opinions, enjoy having them challenged and change your mind when someone proves you wrong. New domains pull you in. You enjoy the moment before the pieces fit - when there’s still something to untangle and a workaround everyone else has stopped noticing. Your career may have zig-zagged across teams or industries because unfamiliar problems are hard to resist. You’re curious without needing to be the expert in the room, and confident enough not to pretend. You listen to the people who know the territory, learn fast and can’t resist the question that turns the problem on its head. You bring energy without performing. You don’t have to be a jazzy extrovert, but you’re lit up by what’s possible and that pulls people along. Requirements What you’ll do The work is varied enough that no neat list will stay true for long! You could find yourself: Working with our GTM innovation squad to explore how agentic approaches could change product-led growth - building, deploying and testing the most promising ideas. Embedding with our legal team to rethink commercial contract review, building something that saves lawyers hours and evaluating whether it helped. Reimagining onboarding with our Culture team, taking parts of the process out of Swigger’s heads and building the context layer before anything can run on it. Coaching a leader on how their team could reimagine their world with AI. Sitting with security engineers to understand what a new MCP connector really exposes, and finding the useful path through what is possible, sensible and safe. Whatever the problem, you’ll get close to the work, build and test rather than theorise, and leave behind a working tool, a team that can own it or a much sharper next step. You’ll also raise our whole game - bringing tools, techniques and ways of working we haven’t thought of yet, and making the sharp, opinionated case for them across PortSwigger. Background - open We’re open on where you’ve come from. You don’t need a security background or a conventional route into AI; product, operations, research, engineering, design, enablement, marketing and consulting could all lead here, as could somewhere less obvious. We care more about what you’ve built, how you think and how you work with others. Realistically, you’re likely to be 5-15 years into a relevant career. Long enough to have shipped things that mattered, still close enough to the wor