★ Saved
Micro1 hires in most countries worldwide. Find all AI training jobs open in your country →
Micro1 logo

CPU Performance Architect

micro1 Micro1 Part-timeRemote Typical on Micro1: $20–180/hr
Posted Nov 16, 2026 · ✓ Still open — checked 4 hours ago
Apply now →

Quick facts

Type
Part-time
Location
Remote
Platform
Micro1 — 345 open jobs
Typical pay at Micro1
$20–180/hour · median $68 (335 jobs on our site)
Payout
Deel (twice a month)
Apply
Free to apply — never pay to get AI training work

Required Skills
Instruction-level CPU performance analysis (x86-64 and/or AArch64 assembly)
CPU microarchitecture
Code-speed estimation tools and cost models
llvm-mca
uiCA
IACA
OSACA
compiler scheduling models
Compiler backend optimization
Performance measurement methodology
About micro1
micro1 is a data lab that helps AI labs & Enterprises train foundational models and agents. We provide frontier evaluations and reinforcement learning environments to improve LLM capabilities, as well as contextual evaluations to monitor and improve AI agents in enterprise settings.

Our AI recruiting agent sources and vets domain experts for our data platform that enables rapid production of high-quality training data. Our research spans high-value, specialized domains (legal, medical, finance, coding, etc.), where expert-level data is the hardest to produce and most valuable for advanced model performance. Our pipeline performance system ensures both quality and velocity.

Role Title: CPU Performance Architect

Role Type: Contractor

Location: Remote

micro1 is engaging CPU Performance Architects to contribute to a customer's specialized project focused on realistic CPU cycle modeling and code-speed estimation. This opportunity involves a brief pre-read and a 1-hour remote consultation, critically reviewing methodologies for benchmarking and simulating assembly code performance on modern processor architectures.

Scope of Work

Review provided project documentation and methodology for CPU cycle estimation accuracy and realism.
Evaluate measurement-based modeling setups, focusing on data validity, handling of noisy timing, and test data partitioning (including considerations for earlier chip revisions and held-out samples).
Provide expert analysis of instruction-level performance for x86-64 and/or Arm AArch64 assembly code, referencing execution ports, reorder buffer, register renaming, and out-of-order execution.
Assess use and interpretation of cycle cost estimation tools such as llvm-mca, uiCA, IACA, OSACA, and compiler scheduling models.
Offer actionable feedback on the fairness, validity, and limitations of current benchmarking approaches, with a focus on potential improvements or pitfalls.
Participate in a 1-hour remote consultation to discuss findings, answer technical questions, and advise on next steps.

Preferred Qualifications

10+ years as a specialist in CPU performance, computer architecture, or compiler backend analysis.
Expert-level proficiency with instruction-level analysis of x86-64 and/or AArch64 assembly, especially in performance measurement contexts.
Hands-on experience with cpu cycle estimation and modeling tools: llvm-mca, uiCA, IACA, OSACA, or equivalent.
Deep understanding of core microarchitectural features (execution ports, out-of-order execution, register renaming, reorder buffers).
Strong background in compiler backend optimization and scheduling models.
Excellent written and verbal communication skills, with a focus on clear, actionable explanation and feedback.
Direct experience in pre-silicon performance modeling and/or Arm AArch64 architecture is a strong plus.
Email

By clicking any of the "Continue" buttons below, you acknowledge and agree to our Terms and Privacy Policy for candidates.

Continue
or
Continue with Google
Continue with LinkedIn

Have any questions? See FAQs

Refer and Earn

micro1 recruits engineers and domain experts for AI labs using its AI interviewer. Once vetted, contractors work on human-data projects such as writing and checking model outputs or building coding evaluations.