Software Engineer, AI
Location: San Francisco, CA (Preferred)
The role
You’ll be a crucial member of the team building Bezi, the game development assistant that leverages complete understanding of a project and AI to help with the repetitive parts of building a game.
As a Senior Software Engineer on our small, high-impact AI team, you'll be instrumental in designing and building the agent pipelines, AI workflows, and backend systems that power Bezi. Working closely with the rest of the team, you'll contribute across the full AI stack, from prompt engineering and retrieval systems to evaluation frameworks and API design, shipping features that help game developers move faster.
Bezi has real-time understanding of the project (codebase, assets, scene graph, components, and more) to provide accurate and custom assistance. It can help with coding and script writing, debugging and optimization, and technical explanations.
The impact you'll have:
Architect and build the core agent systems that power Bezi's game development assistant, directly shaping how developers interact with AI in their workflow.
Design and implement evaluation and testing frameworks to ensure reliable, high-quality AI outputs across diverse game development scenarios.
Own backend systems (APIs, queues, and integrations) that connect Bezi's AI capabilities to real-time project context.
Drive technical decisions around prompt engineering, retrieval (RAG), and tool use / function calling across multiple LLM providers.
Shape AI strategy and architecture as an early member of the team, with significant influence over the product's technical direction.
What you bring:
5+ years of professional software engineering experience, with strong backend fundamentals.
Excellent TypeScript skills and hands-on experience building production systems with Node.js.
Intermediate to advanced knowledge of AI engineering concepts, including prompt engineering, retrieval-augmented generation (RAG), and tool use / function calling.
Professional experience working with AI coding agents such as Claude Code, Codex, or Cursor. You understand how these tools work and how to build with (and for) them.
Prior professional experience working on LLM-powered agents preferred. You've built, shipped, or maintained agentic AI systems.
Comfort working across multiple LLM providers and evaluating tradeoffs between models.
A bias toward ownership and shipping. You thrive in fast-paced, early-stage environments where you can take projects from idea to production.
Nice to have:
Experience in game development or familiarity with game engines, scene graphs, and real-time rendering pipelines.
Experience designing evaluation or testing frameworks for non-deterministic AI outputs.
About Bezi
We empower people building games. We’re a group of technical, first-principles thinkers. Every member of the team is trusted, and expected, to deeply understand the people we build for and find creative ways to improve their lives and work through Bezi. Our organization is flat; everyone is trusted to pursue and validate ideas and every idea is supported and debated equally.
We believe the best work happens when teams know each other. Bay Area team members gather in our San Francisco office three times a week and remote teammates join us in person multiple times a year.
How to apply
Send your resume and a short note about a relevant project you led to careers@bezi.com. If there’s a potential fit, our team will respond as soon as possible.