About the role
About
the role You'll own the core AI systems that power Gamma: the models, prompts, and pipelines behind text, image, and layout generation. With over 1 million AI-generated presentations and 6 million AI images created every day, this work operates at massive scale. Your job is to elevate quality, evaluate new frontier models, and push into new capabilities and modalities. This role is about productizing existing foundation models, not training new ones. You'll focus on prompting, evaluating, and fine-tuning for maximum performance across Gamma's product surface. You'll also launch new modalities like voice and video, build the evaluation infrastructure to measure what "good" looks like, and own uptime, latency, and cost across our AI stack. You'll work closely with engineering and product to ship improvements that millions of users feel immediately. You'll thrive here if you're a tinkerer who loves pushing foundation models to their limits and you're equally comfortable writing prompts and writing production code. If you get excited about mixing prompt engineering with software engineering to unlock new AI capabilities, this is your role. Our team has a strong in-office culture and works in person 4–5 days per week in San Francisco. We love working together to stay creative and connected, with flexibility to work from home when focus matters most.
What you'll do Own Gamma's LLM and image prompts, measuring and continuously improving quality at scale across text, layout, and visual generation Develop complex prompts for new features using AI JSX, balancing creativity with reliability Build evaluation frameworks for prompts and models, combining quantitative metrics with qualitative feedback to create better test sets Drive the AI roadmap based on quality gaps, constantly evaluating new frontier models and methods Curate datasets for fine-tuning open-source models and launch new modalities like voice and video Own analytics, tracking, uptime, latency, and cost across Gamma's AI infrastructure What you'll bring Proven track record pushing foundation models to their limits, with hands-on experience building and evaluating prompts at scale Proficiency in TypeScript and Python, with a software engineering foundation that lets you move fluidly between prompt engineering and production development Strong data instincts: experience writing evals, designing metrics, and translating qualitative feedback into measurable, actionable improvements Self-sufficient in gathering and cleaning data to inform prompt improvements and model evaluations Experience with modern LLMs and image models such as Flux and Imagen (
