Founding Member of Technical Staff

Function•New York City, NY

About The Position

Function is building a platform that enables developers and enterprises to run AI inference on-device. The platform consists of a general-purpose Python compiler and client SDKs that developers use to run compiled functions in their applications. Engineering work spans compiler infrastructure, accelerated computing, and model sourcing, and the areas are closely intertwined, so this role may involve work across all of them. The compiler translates Python code to native code and cross-compiles it for desktop, mobile, web, and cloud. The team also develops and benchmarks optimized kernels for CPUs, GPUs, and neural processors, and compiles popular open-source models for the company’s Explore page. The role offers broad impact, production ownership, customer interaction, and influence over platform design, with work ranging from high-level Python to low-level assembly and LLM-assisted low-level code generation.

Requirements

  • Strong curiosity about high-performance computing and systems-level programming.
  • Ability to write highly performant, safe, and correct C++ and/or Rust code.
  • Ability to inspect emitted object code using tools such as binutils and Ghidra, and perform detailed benchmarking.
  • Experience with platform-specific SIMD and GPGPU offerings, including Vulkan compute, Metal, CUDA, or WebGPU.
  • Experience using high-level inference libraries such as ONNXRuntime, GGML, or MLX.
  • Expertise in Python.

Responsibilities

  • Build and maintain native code generation, implementing Python language features in C++23 and Rust.
  • Design and build pipelines for high-performance native code generation in C++ and Rust, supporting tensor operations, LLM inference, and other operations needed for popular AI models.
  • Maximize compute utilization by leveraging SIMD and GPGPU instruction sets across Android, iOS, macOS, Linux, WebAssembly, and Windows.
  • Contribute across compiler infrastructure, accelerated computing, and model sourcing as needed.
  • Push work to production, communicate with customers, and influence the design of the platform.
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service