Inference Engineer
Inference

The role

We're looking for engineers who can bridge the gap between ML research and high-performance inference.

You'll work across our inference engine and model conversion toolkit, implementing new model architectures, supporting new modalities, writing optimized kernels, and building a wide range of features such as function calling and batch decoding.

This role is ideal for someone who reads papers for fun, enjoys writing high-performance code, and gets excited about constant learning.

Nobody knows everything. We'd rather you know one area deeply than everything superficially. If you're good at least in a couple of these areas, you're a great fit:

JAX / Equinox / Pallas stack

Rust systems programming with a focus on developer experience

Writing Metal / Vulkan kernels

Neural codecs and voice model architectures

Trellis-based quantization approaches

Advanced speculative decoding methods, such as EAGLE

Deep understanding of Transformer / SSM / Diffusion / Vision language models

Benchmarking inference performance and model quality

Strong linear algebra, optimization methods, and probability theory

And of course, basic engineering skills, we will ship a lot of code 🙃

We welcome applications from students and early-career engineers. If you've participated in projects that demonstrate systems thinking and ML understanding, we want to hear from you!

About us

Mirai is building the on-device inference layer for AI.

We enable model makers and product developers to run AI models directly on edge devices, starting with Apple Silicon and expanding to Android and beyond.

Our stack spans from low-level GPU kernels to high-level model conversion tools.

We're a small team obsessed with performance, working at the intersection of systems programming and machine learning research.

Why us?

Founded by proven entrepreneurs who built and scaled consumer AI leaders like Reface (300M users) and Prisma (100M MAU).

Our team is small (16 people), senior, and deeply technical. We ship fast and own problems end-to-end.

We’re advised by a former Apple Distinguished Engineer who worked on MLX, and backed by leading AI-focused funds and individuals.