Product ML Engineer

Role summary

We're looking for a Product ML Engineer to join one of our product experience teams — SMB, Scaler or Enterprise — and own the ML side of short, high-impact projects.

You'll work on focused ML problems with a clear product outcome, sometimes fine-tuning one of our models, sometimes integrating the best externally available model, but always choosing the fastest and most effective path to customer value.

You'll bring the technical depth of an Applied Scientist while operating with the pace and pragmatism of a product engineer. You'll own the journey from framing the problem through to shipping, measuring and iterating on the feature in production.

This is not a research role. Your time horizon is weeks rather than quarters, and success is measured by what reaches users and the impact it creates.

About the role

• Embed directly within a product team. Work closely with the SMB, Scaler or Enterprise team on focused ML problems tied to clear customer and business outcomes.
• Own ML projects end to end. Frame the problem, choose the approach, execute the plan, ship the feature and monitor its performance in production.
• Make the build vs. buy decision. Determine when to fine-tune or build a model yourself versus integrating an external model, API or existing library.
• Move from idea to production quickly. Work on projects scoped in weeks rather than quarters, prioritising rapid iteration and customer impact.
• Work closely with the ML team. Leverage shared infrastructure, evaluation tooling and expertise while remaining embedded in the product team's priorities and cadence.
• Stay close to users. Get rapid feedback from customers, monitor how people use what you've shipped and use those insights to drive the next iteration.
• Balance technical depth with pragmatism. Know when deeper modelling work is necessary and when an existing solution is already good enough to ship.
• Write production code. Work directly in the product team's codebase rather than operating primarily through research notebooks and experiments.
• Own quality in production. Build sound evaluations, monitor performance and investigate failures once the feature is live.

• You've shipped ML into production. You have 2+ years of hands-on experience building and shipping ML-powered products, with strong experience in PyTorch or equivalent frameworks such as TensorFlow or JAX.
• You have applied-science depth. You can fine-tune vision or diffusion models, design meaningful evaluations and debug training runs when necessary.
• You're a pragmatic builder. You care about solving the customer's problem rather than proving that your own model is the best approach. You'll happily use an external model, API or existing library when that's the fastest route to value.
• You're product-minded. You measure success through shipped features, user behaviour and business outcomes rather than papers or benchmarks alone.
• You're comfortable in production codebases. You can move between experimentation and production engineering, writing maintainable code that lives inside a real product.
• You have strong ownership. You're comfortable taking an ambiguous ML problem, deciding what matters, choosing an approach and driving it through to production.
• You iterate quickly. You enjoy short feedback loops and would rather ship something useful, learn from users and improve it than spend months optimising something that hasn't reached production.
• You're curious and collaborative. You can both explain your expertise and learn from the deep technical expertise around you. You value knowledge sharing and humility.
• You're fluent in English. French isn't required.

Nice to have: experience shipping generative image models or diffusion models into production; experience evaluating third-party AI APIs against internal alternatives; experience working in a fast-growing startup or directly within a product team.

• Screening with an ML Researcher / Engineer
• Home technical assignment and review with the team
• Culture fit interview and meet the team
• Reference check and offer

Support: We value diversity and aim to create an inclusive experience for all applicants. Please let us know if there's anything we can do to make the process more accessible for you.

• Competitive salary + meaningful equity
• Hybrid working from our Paris headquarters
• Visa sponsorship and relocation support up to €10,000
• MacBook Pro
• Private healthcare
• Regular company off-sites and team retreats
• International English-speaking team
• Language lessons in English and French

Founded in 2019 and part of Y Combinator’s Summer 2020 cohort, Photoroom is the leading AI photo and video solution for e-commerce, helping businesses sell at first sight.

Backed by Series B funding, we’ve reached 300M+ downloads and now process around 7 billion images a year for businesses of every size, from solo sellers to global brands and marketplaces like Amazon, Decathlon, and Depop. We serve them through our mobile app, web platform, and API, with tailored solutions for enterprise catalogs.

We’re an international team of 130+, with offices in Paris and New York. We take ownership of what we build, move quickly to learn what works, and speak candidly, united by our ambition to create exceptional impact for e-commerce businesses worldwide.