AI Inference Performance Scientist

Cerebras Systems; Binangonan; On-site; PHP 600,000 - 1,100,000; Full time

Job summary

Cerebras Systems is hiring for the Inference Core Platform group to deliver the world’s fastest AI inference. You’ll work on end-to-end telemetry, benchmarking infrastructure, and performance analysis for large-scale systems on the Cerebras Wafer-Scale Engine.

Join a team that drives bring-up of core inference capabilities, improves speed and throughput, and collaborates across engineering and leadership to ship features with measurable impact.

Qualifications

• Bachelor's or Master's in Computer/Systems Engineering or related field.
• Proficiency in Python and/or C++.
• Proven experience building and scaling automated infrastructure.
• Strong background in throughput and performance optimization for large-scale systems.
• Excellent problem-solving and analytical mindset.
• Ability to dive deep into new domains.
• Experience working in a fast-paced, collaborative environment.

Responsibilities

• Design end-to-end telemetry systems across the software stack for end-to-end observability of inference performance.
• Architect, build, and scale automation that analyzes performance data to inform decisions.
• Investigate system behavior to identify bottlenecks and deliver actionable performance improvements.
• Collaborate with Core Platform teams to define testing methodologies ensuring peak performance of inference features.

Skills

Python

C++

Automation infra

Performance optimization

Problem solving

Deep dive analysis

Collaboration in fast-paced env

Education

Bachelor's or Master's in Computer/Systems Engineering

Job description

Cerebras Systems is hiring for the Inference Core Platform group to deliver the world’s fastest AI inference. You’ll work on end-to-end telemetry, benchmarking infrastructure, and performance analysis for large-scale systems on the Cerebras Wafer-Scale Engine.

Join a team that drives bring-up of core inference capabilities, improves speed and throughput, and collaborates across engineering and leadership to ship features with measurable impact.