Teams at Modular

Develop and optimize Mojo kernels and implement support for new accelerator architectures across Modular’s AI software stack. Build portability infrastructure, integration tests, and debugging workflows; collaborate with hardware vendors to triage platform issues. Engineers will study ISAs, memory hierarchies, and vendor toolchains while contributing to compiler, model-serving, and inference optimization efforts.

Design and implement Mojo language features, compiler optimizations, and multi-target compilation for CPUs, GPUs, and AI accelerators using MLIR and LLVM. Collaborate with kernel authors, hardware engineers, researchers, and the open-source community to improve language design, code generation, performance, developer experience, and the broader Mojo SDK.

Own and evolve MAX Framework APIs and developer experience for AI inference and training. Design scalable APIs, programming models, distributed execution primitives, and developer-facing abstractions across Python and Mojo-adjacent surfaces. Lead RFCs and technical specifications, build reference implementations, establish API quality standards, and collaborate with compiler, runtime, kernels, serving, documentation, and DevRel teams.