Chief of Staff

About Refine

A spreadsheet error in one of the most influential economics papers of the decade helped justify austerity across Europe and the US. Millions lived with the consequences. The error was only found years later, by sheer luck, long after the damage was done.

Flawed analysis shapes policy, investment and lending every day, costs billions every year, and mostly goes unnoticed. AI makes it worse: rigorous-looking work is now cheap to produce.

Refine exists to catch these errors before anyone acts on them. We're building the infrastructure that verifies high-stakes work; think CI/CD for everything beyond code. It reviews a document like a team of experts, checking the maths, data, logic and argument, in minutes.

• The best AI review system. Refine 5 beats frontier models such as GPT-6 Astra in over 90% of blind head-to-heads across ten fields, from economics to biology. It combines the top frontier models at hundreds of times normal compute, so no one model's blind spots get through. We tune it on documents seeded with the errors frontier models miss.

• The de facto AI reviewer for economics. Econometrica and the AER run every paper they publish through Refine. It's used at 24 central banks and by researchers at Oxford, MIT, Stanford, Yale and Cambridge; Cambridge, UCLA and NYU provide it to theirs.

• Beyond academia. 29,000+ reviews sold at 1,000+ institutions, including litigation consultancies, hedge funds and companies bidding on public tenders.

• Enterprise-grade. ISO 27001 certified, SOC 2 Type 1 (Type 2 in progress). On the web and inside ChatGPT, Claude and Codex.

• Where we are. Seed stage: fewer than 10 people, $5–10M raised, and revenue in the $1–10M range.

The role

You'll be the founders' eyes and ears, working directly with the two of us - Yann Calvo Lopez (CEO) and Ben Golub (Chief Scientist, formerly a tenured economics professor at Harvard) - on whatever needs doing that week. Our advisors include OpenAI co-founder John Schulman.

Short term, you take execution work off our plates: the kind that needs context on the whole company. Over time, you build the engineering and business context to grow into a managerial role.

Most of it you'll never have done before. You'll use AI to do it anyway, and do it well. It's not an easy job; there's a lot of grunt work, and the pace is very fast.

A typical week

• A founder has an idea; you run it end to end. Say a hedge-fund chairman who publishes research. You pull his recent papers, run them through Refine, pick the finding worth his time, and steer Claude to a message we'd be proud to send. Then you ping us.

• Prepare the meetings that matter. Context first: who's in the room, what they care about, what we want out of it.

• Make every call forwardable. The deck or one-pager our contact sends to their boss, built from the transcript, same day.

• Follow up on everything. Every next step chased, every deal current in the CRM. Nothing goes cold.

• Run Refine on the documents that matter, and turn the results into something a prospect, partner or investor can act on.

• Make things happen with the team. You know what everyone's working on, and you unblock it.

• Whatever else needs doing. Some of it is grunt work; it all matters.

How we run

Our company memory lives in repositories that AI agents keep current, with a source and a date on every entry. Claude is connected to the tools we use - Slack, Drive, Notion, HubSpot, PostHog and more. You get full-company context from day one, and so does your AI.

Who you are

• Highly intelligent, and a fast learner. You get to the bottom of a new field, tool or problem in days.

• AI-native, with judgment. Claude Code, Codex, Claude Cowork or similar, every day, for real work. You know when, what and how to delegate to AI, and you check what comes back. Show us something real.

• Dedicated; ideally obsessive. Loose ends bother you. You keep going until it's right, and you follow up without being asked.

• A clear writer. Emails, one-pagers and decks that sceptical experts actually forward.

• At home in chaos. Self-sufficient when things get fast and messy. Low ego; we're very direct with each other.

• Fluent in English.

Title and years of experience don't matter if this describes you.

Nice to have: an academic background (a research degree, or familiarity with how academia works); experience in a startup, consulting or a founder's office; Spanish or another European language.

Probably not for you if

• You want a narrow job description. This one is whatever needs doing.

• You only want the strategy. There's a lot of grunt work.

• You want a steady pace. It's very fast here.

How we work

• AI-native throughout. One person with the right setup does what used to take a team.

• Small team, strong engineering. Everyone executes, founders included, and engineering ships fast.

• Autonomy. You choose what to work on, by impact.

• Speed. We build at the speed of AI.

• Remote-first. We're spread across Europe and the US. We work async by default, with regular team syncs, frequent customer calls and the occasional in-person offsite.

• Respect for users. They're sceptical experts; we win them with substance, not hype.

• Meritocracy. Talent, hard work and results are rewarded.

Compensation: competitive.

How we hire

1. Written answers to the questions below.

2. An exercise close to the real work.

3. Conversations with the founders.

4. A paid trial: one or two days working with us on real work.

How to apply

Look around refine.ink, then send your CV or LinkedIn with short answers to:

1. Tell us about something you'd never done before that you did well with AI. What did you hand to AI, what did you keep, and how did you know it was right?

2. What did you get obsessed with, and how far did you take it?

How you think matters more than polish.