# Developer Advocate, MAX Inference & Serving at Modular

- Company: Modular
- What the company does: The unified AI inference stack - from custom GPU kernels to production cloud serving on NVIDIA and AMD. 2x performance. Top open models. Open source stack. Backed by General Catalyst, Greylock and GV.
- Company website: https://www.modular.com/
- Type: Startups (AI role)
- Level: Mid level
- Location: United States - Remote
- Work setup: Remote
- Posted: 2026-08-07
- Apply by: 2026-10-08
- Apply: https://jobs.gem.com/modular/am9icG9zdDr8mLYVA244cbADr5tGwP5M
- Page: https://www.1752.vc/careers/jobs/modular-developer-advocate-max-inference-and-serving/

## About the role

We are looking for a Developer Advocate to evangelize the MAX Platform's inference and serving capabilities with our user base and developer community. This involves creating technical content such as user guides and blog posts as well as giving talks at conferences, leading workshops, all with the goal of enabling our community of builders deploying models in production. Join our world-leading product team and be part of redefining how AI infrastructure is built and deployed.

## What they're looking for

- 3-5 years of professional engineering experience, with at least some of it spent deploying or operating ML systems in production. You have run inference workloads yourself, not just written about them
- Strong Python and systems programming experience in C++, Rust, Mojo, or CUDA is a significant advantage, especially if you have profiled and optimized GPU code
- Comfort with the deployment surface around serving: containers, Kubernetes, GPU drivers and runtimes, and the usual ways a cluster refuses to cooperate
- You benchmark rigorously. You know why one number is not a result, how to control for warmup and batch size, and when a comparison is not apples to apples
- You write well about technical work, and you have a portfolio to show it: blog posts, tutorials, videos, docs, or courses. You explain complex ideas without losing precision, and you cut the filler
- You learn new tools fast and turn them into accurate content quickly. A feature ships Tuesday, your tutorial goes out Thursday and the code runs