AI Applied Scientist

About the role

Figma is actively searching for an AI Applied Scientist to spearhead the development of advanced machine learning technologies that directly enhance our design platform and redefine how teams create. In this role, you will own the research, experimentation, and production implementation of intelligent features that automate complex design tasks and unlock new creative possibilities for users. You will conduct fundamental and applied AI research to push the limits of current technology, ensuring that Figma remains at the forefront of innovation in the AI design space. A significant portion of your work will involve building generative AI models using supervised finetuning, reinforcement learning, prompt optimization, and synthetic data generation to solve real-world design challenges. You will partner closely with product and infrastructure teams to integrate these AI-powered capabilities seamlessly into Figma, ensuring they are robust, scalable, and user-centric. Another core responsibility is to translate user feedback and product hypotheses into precise technical requirements by collaborating deeply with product managers and engineers. You will also design and implement rigorous evaluation frameworks to continuously assess and improve the quality, safety, and effectiveness of AI features once they are in production.

Key facts

What you'll do

• Conduct fundamental and applied AI research to push the limits of current technology and identify novel approaches for design problems.
• Build generative AI models using supervised finetuning, reinforcement learning, prompt optimization, and synthetic data generation to create reliable and intelligent features.
• Partner with product and infrastructure teams to integrate AI-powered capabilities into Figma, ensuring seamless performance and user experience.
• Translate user feedback and ambiguous product goals into clear technical requirements by collaborating with product managers and stakeholders.
• Design and establish comprehensive evaluation frameworks to assess and improve the quality, accuracy, and safety of AI features throughout their lifecycle.
• Implement and iterate on machine learning models using Python, ensuring that solutions are production-ready and maintainable within large codebases.
• Lead experiments to measure the impact of AI features on user behavior and creative outcomes, using data to drive product decisions.
• Contribute to the development of best practices for prompt engineering, model tuning, and synthetic data generation within the organization.
• Collaborate with cross-functional teams to identify opportunities where AI can enhance prototyping, collaboration, and design system management.
• Stay up to date with the latest advances in generative AI, large language models, and reinforcement learning to bring cutting-edge ideas into the product.
• Mentor and influence senior engineers by sharing technical insights and guiding the implementation of complex AI solutions.
• Define and track key metrics to evaluate the success of AI features, ensuring they meet user needs and business objectives.
• Work within a fast-paced environment, balancing long-term research with the urgent demands of product delivery.
• Ensure that all AI implementations adhere to principles of transparency, fairness, and user trust.

Requirements

• 4+ years of experience in Generative AI, with a focus on applying these techniques to real products.
• 6+ years of experience in machine learning, natural language processing, or computer vision, demonstrating deep technical expertise.
• 5+ years of software engineering experience using Python, C++, Java, or R, with a strong portfolio of shipped systems.
• Proven ability to build generative AI features in production environments via fine-tuning and prompt engineering, with measurable results.
• Proficiency with deep learning frameworks such as PyTorch, JAX, or HuggingFace, and experience deploying models at scale.
• Experience training LLMs using reinforcement learning techniques like DPO, PPO, or RLVR (GRPO/DAPO), including tuning hyperparameters and analyzing outcomes.
• Ability to communicate and collaborate across functional teams, translating technical concepts for non-technical stakeholders.
• Strong problem-solving skills and the capacity to debug complex issues in model behavior, data pipelines, and integration points.
• Commitment to writing clean, efficient, and well-documented code that can be maintained by other engineers.
• Willingness to participate in on-call rotations to support production AI features as needed.

Nice to have

• Experience planning multi-year roadmaps that align short-term projects with long-term strategic goals for AI within the product.
• Background in mentoring or influencing senior engineers through technical leadership and knowledge sharing.
• Expertise in large-scale, distributed AI training, including optimization strategies for cost and time efficiency.
• Experience with reinforcement learning from execution feedback for coding tasks or non-verifiable rewards, particularly in agent-based systems.

Practical notes

Compensation is determined based on individual qualifications, market demands, and location. Benefits include equity, health coverage, retirement contributions, parental leave, and flexible PTO. Candidates must keep cameras on during video interviews and attend in-person onboarding if hired. Applicants requiring accommodations for the hiring process may contact accommodations-ext@figma.com.