Senior Machine Learning Scientist

Role:

Responsibilities:

• Own end-to-end machine learning solutions, from problem framing and data strategy through to modelling, deployment, and iteration in production
• Design and build scalable ML systems to detect fraud, abuse, and policy violations in user-generated content across text and multimodal domains
• Lead the development and application of LLM-based approaches, including model selection, fine-tuning, evaluation, and failure analysis
• Define and drive experimentation strategy, including offline evaluation and online testing, to rigorously measure impact and inform product decisions
• Work in ambiguous, evolving problem spaces, proactively identifying new risks and shaping detection strategies in partnership with Trust, Policy, and Product
• Collaborate cross-functionally to translate business and safety goals into effective, production-ready ML systems, influencing trade-offs and priorities
• Communicate clearly and effectively with both technical and non-technical stakeholders, articulating approaches, trade-offs, and impact

Qualifications:

• Proven track record of designing, deploying, and iterating on machine learning systems that deliver measurable impact in production environments
• Strong foundation in machine learning and deep learning, with hands-on experience using frameworks such as PyTorch and modern architectures (e.g. Transformers, large language models)
• Experience applying ML in real-world, noisy, and adversarial domains, such as trust & safety, fraud, or abuse detection
• Proficiency in Python and experience writing production-quality code, with a solid understanding of data pipelines, model training workflows, and MLOps practices
• Demonstrated ability to own problems end-to-end, operate in ambiguous environments, and make pragmatic technical decisions
• Strong collaboration and communication skills, with the ability to influence cross-functional partners and stakeholders
• Experience building ML systems for tr ust, safety, fraud, or policy enforcement use cases
• Hands-on experience fine-tuning, evaluating, or deploying large language models in production settings
• Experience with multimodal modelling (e.g. text + image)
• Familiarity with human-in-the-loop systems or moderation workflows
• Experience with Databricks, PySpark, or large-scale data processing systems
• Additional Information
• Health + Mental Wellbeing
• PMI and cash plan healthcare access with Bupa
• Subsidised counselling and coaching with Self Space
• Cycle to Work scheme with options from Evans or the Green Commute Initiative
• Employee Assistance Programme (EAP) for 24/7 confidential support
• Mental Health First Aiders across the business for support and signposting
• 25 days of annual leave with the option to carry over up to 5 days
• Impact hours: Up to 2 days of additional paid leave per year for volunteering
• Fully paid 4-week sabbatical after completion of 5 years of consecutive service with Depop, to give you a chance to recharge or do something you love.
• Flexible Working: MyMode hybrid-working model with Flex, Office-Based, and Remote options role-dependent
• All offices are dog-friendly
• For birth parent: 20 weeks of paid parental leave for full-time regular employees
• For non-birth parents: 12 weeks of paid parental leave for full-time regular employees
• IVF leave, shared parental leave, and paid emergency parent/carer leave
• Twice-yearly development chats and yearly performance reviews
• Learning budget
• Upskilling our employees with company-wide training workshops, materials and resources
• Life Insurance (financial compensation of 3x your salary)
• Pension matching up to 6% of full base salary with Aviva
• In-office Depop Shop (that’s free!) and a packing station with free delivery.
• Special milestones are celebrated with gifts and rewards!

Bonus points: