Lead Developer Advocate
-
Software Development
Job Description & How to Apply Below
Parasail is the multi-silicon inference cloud. It matches each workload to the right hardware across NVIDIA, AMD, Positron, and other next-generation accelerators and tunes the kernels underneath, then battle-tests the results across trillions of tokens a day. Every optimization ships to every customer, so every team gets performance tuned for their workload from one endpoint in minutes. Parasail is headquartered in San Mateo, California, and backed by Kindred Ventures, Turing Capital, Basis Set and others.
Therole
We’re looking for our first developer advocate to be Parasail’s technical voice on inference, from open weight models and the hardware they run on to the applications built on top.
You’ll run your own experiments on how inference behaves in production, write up what you find, present it at conferences and lead workshops for developers. Day to day, you’ll work with our perf and platform engineers and represent Parasail with the model labs and hardware partners we build with. The role starts as an individual contributor, with room to build and lead a team as it grows.
Whatyou’ll do
Advance the technical conversation. Do original work on how inference behaves in production: serving engines, hardware, latency, cost and quality. Publish the deep dives and references engineers bookmark and cite. Share your work where engineers already are, on X, Linked In and Hacker News, and through Parasail’s own accounts for launches and results.
Take it to the stage. Write CFPs and give talks at the conferences and meetups where AI engineers, infrastructure teams and model builders show up. You’ll also run workshops and go on podcasts and panels.
Work with the labs and the chip makers. Launch new models with open weight labs. Show developers what new silicon can do with partners like AMD, Positron and d-Matrix. Build integrations with the frameworks and tools developers already use. Align with the labs and chip makers on how to benchmark, and build joint benchmarks.
Engineers learn from Parasail’s technical work, cite it and share it.
Parasail is a regular presence on stage at the conferences and meetups where AI engineers show up.
An open benchmarking or research effort, shaped together, that other teams use.
Joint launches and technical content with open weight labs and hardware partners.
Our engineers co-author and speak regularly, and there’s a clear plan for growing the team if the motion is working.
You’ve done technical work in public, like conference talks and writing that engineers cite, whether your background is developer advocacy, ML education, research or engineering.
You build with LLMs in Python, can read serving engine code, and can explain the GPU-level tradeoffs behind latency, throughput and cost.
You can make hard systems ideas clear without dumbing them down, on stage and in writing.
You’ve contributed to or maintained open source that developers use.
You know people across the AI ecosystem, or you know how to meet them fast.
You’d rather build a function than inherit one, and you’re comfortable working without a playbook.
Time in vLLM or SGLang, or on accelerators beyond NVIDIA
Published evals or benchmark methodology for LLM systems
Accepted talks at major AI, ML or systems conferences
A following on X, Linked In or elsewhere, earned through your technical work
A research or teaching background in ML or systems
A CEO who has built inference since 2012. Mike Henry founded Mythic and was interim CPO at Groq, and you’ll work with him directly.
Hardware most developers can’t get to. Run your experiments on our multi-silicon fleet: NVIDIA, AMD and new accelerators.
A founding role. You set the…
(If this job is in fact in your jurisdiction, then you may be using a Proxy or VPN to access this site, and to progress further, you should change your connectivity to another mobile device or PC).
View / Apply for Jobs
Matching My Jurisdiction