Full Stack Engineer, AI Accelerator Open Source Ecosystem
Pubblicato il 07-09-2026 - SemiAnalysis in Roma
About SemiAnalysis SemiAnalysis is an independent research and analysis firm specializing in the Semiconductor and AI industries. Our in-depth coverage spans the entire supply chain, from semiconductor fabrication processes to state‑of‑the‑art AI Models, CUDA kernels, and GPU cloud infrastructure. We are recognized as the leading authority on AI infrastructure, with the highest concentration of industry experts within one team, and a deep‑rooted passion for delving into the intricacies.
We’re a global team of over 20 analysts & engineers, each with extensive networks across the semiconductor supply chain and AI ecosystem, publishing industry‑shaping articles while participating in 40+ conferences annually. Our newsletter reaches more than 200 000 subscribers worldwide, including senior management and C‑suite leaders at the leading semiconductor and AI companies.
Some of Industry‑shaping Articles
- InferenceMAX: The world first open inference benchmark that continuous benchmarks performance of popular frontier models
- MI300X vs H100 vs H200 Training:
Extensive Benchmarking & Deep Dive into CUDA & ROCm training developer experience & performance
- Trainium2 Architecture & Networking: Deep Dive into Amazon’s new Trainium2 rack scale system & 3D torus scale up topology
Find out more here:
Position Overview We are seeking a highly motivated & skilled Member of Technical Staff to join our growing engineering team.
Member of Technical Staff will play a crucial role in building & scaling our internal (soon to be open sourced) AI Accelerator Open Source Ecosystem Tracker — a full‑stack Next.js application that tracks pull requests, test parity, CI health, and performance benchmarks across all open‑source projects in the CUDA/ROCm/Trainium/TPU software ecosystem. You will also play a role in writing and preparing weekly digests & technical reports on upstream contribution health, ROCm/CUDA/Trainium/TPU parity trends, and cross‑accelerator benchmark results, with the opportu