Job description
ML4Good • Paris, France
Detailed role description
About ML4GoodML4Good runs intensive, in-person AI safety bootcamps. We help talented people gain the knowledge, motivation and community they need to move into AI safety work. Bootcamps are residential, free to attend, intensive, and bring together people from the same region so they can stay connected afterwards.
We started in 2022 with purely technical bootcamps. We've since added a governance and strategy track and have recently tested a programme targeting legal professionals the European Seminar for Frontier AI and Law. We've run 30 bootcamps with around 600 alumni in total.
We've recently been granted $7.5 million to scale our operations over three years. We plan to scale to 20 bootcamps in 2027 and 30 in 2028, introduce new curricula, and try new formats. These include a Founder Bootcamp for people starting AI safety organisations, an event incubator, and bootcamps targeting different profiles.
The roleYou would co-lead ML4Good, with a focus on new programmes: what we teach and how we teach it. We've roughly split the work like this, but the split will depend on your strengths and on who else we hire.
You would lead on:
- What we teach, which bootcamps we run, and who we target
- Recruiting, training and managing teachers
- Managing curriculum development for existing and new bootcamps
- Designing new programmes
- Strategy for the organisation
Jack, the current director, would lead on:
- Strategy for the organisation
- Managing the programme team
- Process improvement
- The non-curriculum side of new programmes
- Partnerships and alumni engagement
- Hiring for non-teaching roles
You'll both work on strategy. This includes big questions, like whether our current bootcamp model is the best way to reduce AI risk, and smaller ones, like which regions and participant profiles to focus on.
At first, the role will focus more on running programmes while you learn how ML4Good works. You'll take on more of the co-director responsibilities over time.
The teamThe core team is 3-4 people and we also work closely with a team of 6 headteachers, and over 20 teachers. We're growing the team over the next year.
By the new year, ML4Good will also have a formal board, which you would report to alongside the other co-director.
We're a fairly flat organisation. We discuss things openly, and everyone contributes to decisions. Although we expect to add more specialisation, we want to keep this culture as we grow.
We have a base in Paris, and have strong connections with other organisations based in the Paris Hub - CeSIA in particular. However, we are currently a hybrid team. We all enjoy working in person, and hope to spend a significant chunk of 2027 working together.
What you'll do- Teaching teams. Recruit, train and manage head teachers and curriculum developers. Help teaching assistants develop into head teachers.
- Curriculum. Keep our technical and governance curricula up to date. Lead curriculum development for new bootcamps. You don't need to write every session yourself. You should be able to work with experts, think hard about teaching objectives, and judge whether a session works.
- New programmes. Design and oversee new programmes, for example:
- The Founder Bootcamp is a programme we are starting to develop aimed at people who want to start organisations, tentatively running in Spring 2027.
- We're also considering a new biosecurity-focused bootcamp and bootcamps targeting various professional groups
- Strategy. Use your understanding of the AI Safety space to decide where ML4Good should focus. Make big decisions on where we want to go as an org.
- Attending and teaching at bootcamps. You’ll be encouraged to attend bootcamps and events. This may also include teaching, if this suits you.
- General support. As a small organisation, we all step in when needed. This could mean helping with funding applications, interviewing applicants, logistics at bootcamps or taking programme and operations tasks on.
You must have:
- A strong understanding of AI safety. You understand why AI risk is serious and why it's a hard problem, and you have reasonable views on the different approaches people are taking.
- Some high-level understanding of technical AI safety. You don't need to be a technical researcher, but you should be able to explain the main technical and governance agendas in AI safety. We're open to people whose background is mainly in AI governance, strategy, field-building, or technical AI safety.
- Good judgement. This is the most important trait for this role. You make reasonable, well-thought-through decisions, and you can identify talent.
It would help if you have:
- Teaching experience
- Experience designing a curriculum
- Experience managing people
We're also looking for people who:
- Feel a sense of urgency about AI risk
- Take the work, and doing the right thing, seriously
- Strong honesty norms - strong willingness to admit and learn from mistakes
- Are low-ego and collaborative, and listen to others
- Are reliable, and get tasks done
- Care about cost-effectiveness
- Are willing to do boring tasks when they need doing
- Salary: £50,000–£75,000 per year, depending on experience
- Hours: Full-time
- Start date: January 2027 or earlier
- Location: Our team works from a shared office in Paris. We'd like you to move to Paris or spend a significant amount of time here. Time zones matter a lot to our working style, so we strongly prefer candidates who can work hours that overlap with Paris +/-1 hour.
- Visa sponsorship: Available
- Travel: You'll need to travel to some bootcamps. This will be an estimated two per season (=4 per year), the majority, if not all, in Europe, depending on your preference. We'd also like you to be able to travel to AI Safety conferences, and EAGs to represent ML4Good.
The process has four stages:
- Fill in this application form by November 15, 2026 at 11:59 PM (AOE)
- Short work test
- Interview
- Three-day paid trial (mid-December 2026)
If you have any questions regarding this role feel free to reach out to [email protected].
Right: Sticky info cardType
Full-time
Location
Paris (preferred)
Hybrid
Deadline
Nov 15, 2026
Apply now Footer