We're building the creative layer for modern communication. Every month, over a billion people make presentations — but the tools they use to make them haven't evolved in decades. We're changing that, using AI to disrupt a massive market.
📈 Millions of people rely on Gamma to create, teach, and persuade, creating more than 1 million gammas every day.
💻 We see Gamma as the next great workplace tool, combining viral B2C love with a massive B2B opportunity. We believe AI can be a true creative partner: one that understands context, clarity, and taste.
💸 We’ve reached a $2.1B valuation, crossed $100M in annual recurring revenue, and have been profitable since 2023.
💙 We're an imaginative, passionate team who takes our work seriously, but not ourselves. Our culture is warm, a little quirky, and fueled by curiosity.
You'll own the quality of AI across everything Gamma creates. As our Research Engineer, you'll design evaluation frameworks that measure AI output quality, systematically improve production prompts, and fine-tune models to ensure millions of users get exceptional results every time they generate content.
This role sits at the intersection of research rigor and product impact. You'll diagnose failure patterns in AI-generated presentations, docs, and websites, then craft targeted improvements through iterative experimentation. You'll build the tools and workflows that enable rapid testing, validate changes against quality benchmarks, and ensure our AI gets smarter with every iteration. You'll succeed here if you combine deep technical expertise with a research-oriented mindset, comfort working in ambiguity, and an attention to detail that catches dimensions of quality others might overlook.
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.
- Design and maintain evaluation frameworks that measure AI output quality across all Gamma experiences, developing metrics and benchmarks to assess model performance
- Systematically improve production prompts through iterative experimentation, diagnosing failure patterns, crafting targeted improvements, and validating against quality benchmarks
- Conduct rigorous experiments to understand model behavior, analyze results, and derive insights that inform prompt and model improvements
- Build tools and workflows to support rapid experimentation and quality analysis, enabling faster iteration on AI improvements
- Fine-tune models on targeted datasets to improve baseline performance, preventing issues like poor layout choices or low-quality outlines
- Partner with product and engineering teams to ensure AI quality improvements ship quickly and work reliably at scale
- 2+ years working with AI systems, with demonstrated experience shipping production-grade AI products
- Deep hands-on experience with prompt engineering, LLM experimentation, and systematic evaluation of AI outputs
- Strong experimental mindset with the ability to design tests, analyze model performance, and iterate toward measurable quality improvements in ambiguous problem spaces
- Experience with post-training techniques for LLMs including reinforcement learning and supervised fine-tuning
- Exceptional attention to detail and genuine quality obsession, with care for output quality across all dimensions including less visible aspects
- Bachelor's degree in Computer Science, Machine Learning, or a related field, or equivalent hands-on experience with AI research and experimentation (Nice to have)
The base salary for this full-time position, which spans multiple internal levels depending on qualifications, ranges between $180K - $340K plus benefits & equity.
Final offer amounts are determined by multiple factors, including but not limited to experience and expertise in the requirements listed above.
We're building on a full Typescript stack centered around some of the most modern and popular technologies.
- We use our own custom, open-source AI prompting framework, AIJSX. We have a lot of custom tools built in-house, but also new ones like Vercel AI SDK.
- Full end-to-end observability using Datadog, Braintrust, and our own data warehouse
AI generated presentations daily
users around the world
AI images generated daily
LLM tokens processed per month
- You get energy from small teams doing big things.
- You love when design, code, and storytelling overlap.
- You default to action, even when the answer isn’t clear yet.
- You value details, but know when to ship and move on.
- You bring both the spreadsheets and the sparkle, equal parts workhorse and unicorn.
- You believe AI should amplify creativity, not replace it.
- You know kindness and intensity are not opposites.
- You like working with people who care deeply: about their craft, their teammates, and the users on the other side of the screen.
Gamma is full of imaginative, passionate people who take their work seriously but not themselves. The culture is warm, a little quirky, and fueled by curiosity. It’s the kind of place where you’ll debate a pixel on Monday, laugh over someone’s keyboard setup on Tuesday, and ship something remarkable by Friday.
We care about craft, move with intention, and don’t mind getting a little scrappy. It’s fast, creative, and occasionally chaotic — but that’s what makes it interesting.
We asked the team what makes Gamma, Gamma. Here were some of their answers:
- "quirky, inspiring, fun, a little wild in the best way"
- "You can have an idea and just run with it."
- "Everyone’s talented and humble — the mix keeps you sharp."
- "We ship cool stuff, learn a ton, and laugh a lot doing it."
We're a team of dreamers and doers building in beautiful San Francisco 🌉
We're kabbadi enthusiasts, pickleballers, dog herders, woodworkers, keyboard nerds, potters, and more — and we can't wait to meet you!
