P

Positron

Memory-first AI inference silicon for LLMs

Systems Engineer – Mid/Senior/Principal - Experiments & Integration

Hardware Validation & TestingPosted on Nov 14
Full-timeRemote in USA

About the Job

Requirements

  • At least four years of hands-on systems or performance engineering experience.
  • Demonstrated experience running large language model inference workloads end to end and identifying performance costs, including weight loading, memory bandwidth, and host-device synchronization.
  • Strong understanding of performance measurement, including counters, execution traces, profiling, and reporting results with their limits.
  • Hands-on experience with host-side driver and runtime code, and comfort working close to hardware.
  • Ability to act quickly on loosely specified requests, make assumptions explicit, and make results repeatable.
  • Strong Rust skills, or expert-level C/C++ skills and willingness to work in Rust day to day; proficiency in Python; and fluency with Linux, shell scripting, containers, and continuous integration.
  • Ability to communicate direct answers to stakeholders, explain the basis for those answers, and recommend next steps.

Responsibilities

  • Serve as the first responder for simulation and emulation requests from hardware, software, and customer teams, partnering with existing experiments and integration engineers so urgent requests are answered while roadmap work stays protected.
  • Run quick experiments and demonstrations, such as getting a new large language model running on the emulator, profiling where time goes, or trying a configuration requested by a stakeholder.
  • Measure and explain performance using counters, traces, and end-to-end timing, and report the measurement limits and drivers.
  • Help other teams build on the simulators and emulation setup when they need integration support.
  • Convert each investigation into a reusable asset, such as a calibrated model, a repeatable experiment, or a regression test.

Desired Qualifications

  • Experience with hardware emulators such as Cadence Palladium or Synopsys ZeBu, or with FPGA boards.
  • A track record of bringing up new workloads on early or unstable hardware.

About the Company

Positron

Memory-first AI inference silicon for LLMs

Positron develops memory-first AI inference silicon and systems to scale large-language-model serving. Its Atlas systems rely on commodity LPDDR5X memory rather than expensive HBM/CoWoS to boost bandwidth and capacity, enabling more efficient LLM deployment at hyperscale. The company’s next-generation Asimov silicon (tape-out planned for late 2026; production in 2027) pairs compute with very large memory per chip (roughly 288 GB to 2,304 GB) and is designed to be combined into Titan configurations (four to eight Asimov chips) for extremely large models and long context windows. Positron has already deployed Atlas systems in the cloud, including 50+ racks at Oracle Cloud Infrastructure and partnerships with Parasail, Jump Trading, and i3d.net. Its goal is to provide scalable, memory-rich AI inference hardware that lowers cost and increases throughput for serving large language models.

Company size
51-200
Company stage
Series C
Total funding
$1.2B
Headquarters
Reno, NV
Founded
2023

Recently Posted Jobs

Explore more opportunities at Positron

See more jobs