Owns the developer-facing documentation that makes FuriosaAI's software stack — the Furiosa SDK and Furiosa-LLM — usable, from API references and conceptual guides to tutorials, quickstarts, and release notes. This is a docs-as-code role: reading source, running the stack on real hardware, and producing precise, verifiable documentation that keeps pace with a fast-moving compiler, runtime, and inference codebase — backed by automated pipelines that continuously validate that documentation against the live codebase and hardware.
ResponsibilitiesOwns and writes developer documentation for the Furiosa SDK and Furiosa-LLM — API references, conceptual guides, tutorials, quickstarts, and migration guides — translating complex systems behavior into clear, accurate, task-oriented content.
Builds, maintains, and operates the documentation toolchain and docs-as-code pipeline (MDX-based static sites, automated API-reference generation, automated link and sample validation in CI), treating documentation as a versioned, testable artifact that is continuously verified against each SDK release.
Designs and runs automated, hardware-in-the-loop validation pipelines that continuously authors and verifies runnable code samples and end-to-end examples against real RNGD hardware and released packages, catching drift early and ensuring every snippet compiles, runs, and reflects current APIs.
Partners with engineers across diverse teams to capture design intent and surface accurate technical detail, then drives documentation to keep up with release cycles, deprecations, and breaking changes.
Defines and enforces documentation standards — style, terminology, information architecture, and versioning — so content stays consistent and discoverable across the SDK and Furiosa-LLM.
Produces release notes, changelogs, and upgrade and migration guidance that clearly communicate what changed across SDK and Furiosa-LLM versions.
Bachelor's degree in Computer Science or equivalent work experience, with 3+ years writing technical documentation for developer-facing software (SDKs, APIs, systems software, or ML frameworks).
Strong written English with the ability to explain low-level, systems-level concepts precisely and concisely.
Working proficiency in Python and command-line tooling; comfortable reading source code, running build and inference workflows, and writing and validating code samples.
Fluency with Git-based docs-as-code workflows and Markdown/MDX, including running documentation checks in CI.
Solid understanding of deep neural networks (DNNs) and large language models (LLMs) — how they are built, run, and served — sufficient to document inference workflows accurately.
Strong communication skills for cross-team requirement gathering and technical alignment.
Experience documenting ML inference frameworks, compilers, runtimes, or accelerator/GPU software stacks.
Familiarity with LLM inference concepts (serving, batching, quantization, KV cache, distributed inference) and the PyTorch / Hugging Face ecosystem.
Familiarity with GPU Kernel programmings (e.g., CUDA, Triton)
Experience building and operating documentation platforms (e.g., MDX-based static site generators, Fumadocs, Mintlify, Sphinx) and automated API-reference pipelines.
Experience designing information architecture and versioned documentation for software with frequent releases.
Fluency in Python and Rust programming, sufficient to read, write, and validate code samples across the stack.
Skills Required
- Bachelor's degree in Computer Science or equivalent experience with 3+ years writing technical documentation for developer-facing software
- Strong written English and ability to explain low-level, systems-level concepts precisely
- Working proficiency in Python and command-line tooling; comfortable reading source code, running build and inference workflows, and writing and validating code samples
- Fluency with Git-based docs-as-code workflows and Markdown/MDX, including running documentation checks in CI
- Solid understanding of deep neural networks (DNNs) and large language models (LLMs) sufficient to document inference workflows
- Strong communication skills for cross-team requirement gathering and technical alignment
- Experience documenting ML inference frameworks, compilers, runtimes, or accelerator/GPU software stacks
- Familiarity with LLM inference concepts (serving, batching, quantization, KV cache, distributed inference) and the PyTorch / Hugging Face ecosystem
- Familiarity with GPU kernel programming (e.g., CUDA, Triton)
- Experience building and operating documentation platforms (MDX-based static site generators, Fumadocs, Mintlify, Sphinx) and automated API-reference pipelines
- Experience designing information architecture and versioned documentation for software with frequent releases
- Fluency in Python and Rust sufficient to read, write, and validate code samples across the stack
What We Do
FuriosaAI designs and develops data center accelerators for the most advanced AI models and applications. Our mission is to make AI computing sustainable so everyone on Earth has access to powerful AI. Our Background Three misfit engineers with each from HW, SW and algorithm fields who had previously worked for AMD, Qualcomm and Samsung got together and founded FuriosaAI in 2017 to build the world’s best AI chips. The company has raised more than $100 million, with investments from DSC Investment, Korea Development Bank, and Naver, the largest internet provider in Korea. We have partnered on our first two products with a wide range of industry leaders including TSMC, ASUS, SK Hynix, GUC, and Samsung. FuriosaAI now has over 140 employees across Seoul, Silicon Valley, and Europe. Our Approach We are building full stack solutions to offer the most optimal combination of programmability, efficiency, and ease of use. We achieve this through a “first principles” approach to engineering: We start with the core problem, which is how to accelerate.
