Responsibilities
- Design, build, and productionize ML models for Search, Discovery, Ranking, Retrieval-Augmented Generation (RAG), and generative AI features.
- Build and maintain scalable data pipelines to collect high-quality training and evaluation datasets, including annotation systems and human-in-the-loop workflows.
- Collaborate with AI researchers to iterate on datasets, evaluation metrics, and model architectures to improve quality and relevance.
- Work with product engineers to define and deliver impactful AI features across Figma’s platform.
- Partner with infrastructure engineers to develop and optimize systems for training, inference, monitoring, and deployment.
- Explore new ideas at the edge of what’s technically possible and help shape the long-term AI vision at Figma.
Requirements
- 5+ years of industry experience in software engineering, with 3+ years focused on applied machine learning or AI.
- Strong experience with end-to-end ML model development, including training, evaluation, deployment, and monitoring.
- Proficiency in Python and familiarity with ML libraries like PyTorch, TensorFlow, Scikit-learn, Spark MLlib, or XGBoost.
- Experience designing and building scalable data and annotation pipelines, as well as evaluation systems for AI model quality.
- Experience mentoring or leading others and contributing to a culture of technical excellence and innovation.
Nice to Have
- Familiarity with search relevance, ranking, NLP, or RAG systems.
- Experience with AI infrastructure and MLOps, including observability, CI/CD, and automation for ML workflows.
- Experience working on creative or design-focused ML applications.
- Knowledge of additional languages such as C++ or Go is a plus, but not required.
- A product mindset with the ability to tie technical work to user outcomes and business impact.
- Strong collaboration and communication skills, especially when working across functions (engineering, product, research).
Work Arrangement
Remote (Country) — US hubs, remotely in the United States
Additional Information
- This is a full time role.
- Figma values 'Grow as you go' and encourages candidates who are excited about the role even if their experience doesn't perfectly align.
