Job description

Employment Type: Full-Time Work Setting: In-office/Remote Work Location: United States, New York, Mexico, San Francisco, Canada Work Hours: Office hours Find out more here: semianalysis.com 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 cutting-edge AI Models, software, and infrastructure. We are recognized as the leading authority on the semiconductor supply chain, 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 50 analysts, 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. We also offer three core products: • Industry Models – we develop and publish industry models on accelerator shipments, datacentre demand and supply, GPU total cost of ownership, and more. We work with hyperscalers, neoclouds, many of the world’s largest hedge funds, and government agencies. • Core Research – our public equity markets product, geared towards financial investors, distils our deep technical research and knowledge into key insights on technology and product trends. • Consulting and Technical Due Diligence – We conduct custom research and project work to guide key strategic and investment decisions for the largest private equity funds, leading venture capital firms, companies across the AI ecosystem, and government agencies. Position Overview We are looking for a highly motivated member of technical staff to join our engineering team to work on system modelling on 100k+ chip AI clusters. This is an unique opportunity to work on an high-visibility open source projects to architecture & model performance for the next generation of AI chips. If you’re passionate about performance engineering, system modelling, first principles, and want to work at the intersection of hardware and software, this is a rare chance to make industry wide impact. As part of the interview process, you'll complete a paid coding challenge designed to reflect typical daily tasks at SemiAnalysis. Competitive Compensation depending on experience, skillset, location & business needs Responsibilities • Develop an complex model in Python to predict performance (MFU, tok/s/gpu, tok/s/user) on current & future AI chip projects across both frontier LLM training & inference models • Implement modern parallelism & inference strategies to generate Pareto frontier roofline curves for performance on each chip • Develop microbenchmarks across multiple vendors (AMD, NVIDIA, TPU, Trainium) to calibrate the system model to as close to reality as possible • Build BoM estimates to model performance per total cost of ownership and performance per watt Requirements • Strong skills in Python • Deep understanding about disaggregated prefill, wide expert parallelism, tensor parallelism, pipeline parallelism, sequence parallelism, etc. • Experience with frontier MoE training and inference workloads • Knowledgeable about GEMM sizes & proportions of wall time & FLOPs on every major operator in an modern transformer • Intuition about Amdahl principles, arithmetic intensity, weak & strong scaling • Previous experience as an ML engineer or kernel programmer Growth Areas • Develop deep expertise in large-scale system modelling for AI infrastructure, including performance prediction across hyperscale (100k+ chip) clusters • Gain a strong first-principles understanding of how hardware, software, and model architectures interact to determine real-world performance • Build advanced knowledge of parallelism strategies and scaling techniques across frontier LLM training and inference workloads • Strengthen the ability to translate low-level system behavior into high-level performance insights, including cost efficiency and power optimization • Gain hands-on exposure to cutting-edge AI hardware across multiple vendors and architectures, staying at the forefront of industry developments • Contribute to high-impact open source projects and build visibility within the AI and infrastructure engineering community • Take ownership of complex modelling frameworks and drive independent research directions as you grow within the team • Develop the ability to influence technical strategy and decision-making through data-driven insights and performance analysis SemiAnalysis LLC participates in the E-Verify program to confirm the employment eligibility of all newly hired employees. For more information, visit e-verify.gov . • E-Verify Poster: everify.gov • Right to Work Poster: justice.gov

Requirements

Must-haves

Employment Type: Full-Time
Work Setting: In-office/Remote
Work Location: United States, New York, Mexico, San Francisco, Canada
Work Hours: Office hours
Core Research – our public equity markets product, geared towards financial investors, distils our deep technical research and knowledge into key insights on technology and product trends.
Position Overview
Develop an complex model in Python to predict performance (MFU, tok/s/gpu, tok/s/user) on current & future AI chip projects across both frontier LLM training & inference models
Implement modern parallelism & inference strategies to generate Pareto frontier roofline curves for performance on each chip
Develop microbenchmarks across multiple vendors (AMD, NVIDIA, TPU, Trainium) to calibrate the system model to as close to reality as possible
Build BoM estimates to model performance per total cost of ownership and performance per watt
Strong skills in Python
Deep understanding about disaggregated prefill, wide expert parallelism, tensor parallelism, pipeline parallelism, sequence parallelism, etc.
Experience with frontier MoE training and inference workloads
Knowledgeable about GEMM sizes & proportions of wall time & FLOPs on every major operator in an modern transformer
Intuition about Amdahl principles, arithmetic intensity, weak & strong scaling
Previous experience as an ML engineer or kernel programmer
Growth Areas
Develop deep expertise in large-scale system modelling for AI infrastructure, including performance prediction across hyperscale (100k+ chip) clusters
Gain a strong first-principles understanding of how hardware, software, and model architectures interact to determine real-world performance
Build advanced knowledge of parallelism strategies and scaling techniques across frontier LLM training and inference workloads
Strengthen the ability to translate low-level system behavior into high-level performance insights, including cost efficiency and power optimization
Gain hands-on exposure to cutting-edge AI hardware across multiple vendors and architectures, staying at the forefront of industry developments
Contribute to high-impact open source projects and build visibility within the AI and infrastructure engineering community
Take ownership of complex modelling frameworks and drive independent research directions as you grow within the team

Nice-to-haves

None flagged.

Application prep

Use the role’s language and application details while tailoring your materials.

Keywords to mirror

aipythonllmml

Platform: careers-page
Effort: medium