Software QA Engineer - Data Analysis and Test Operation

Job Title

Software QA Engineer - Data Analysis and Test Operation

Role Summary

Work on the Solution QA team to validate FuriosaAI NPU boards, firmware, drivers, runtime, SDK and model-serving solutions across development, production, and customer environments. Operate and run test campaigns, analyze test and performance data, and identify defects, regressions, and environmental causes.

The role focuses on test planning, controlled execution, data-driven anomaly analysis, defect reproduction, and concise quality reporting to engineering and release stakeholders.

Experience Level

Mid-level (Level - Mid-Career). Years of experience not specified.

Responsibilities

Plan and execute software and system evaluations, operate tests under controlled conditions, analyze results, and report quality findings.

• Define evaluation scope, test objectives, preconditions, procedures, expected results, and pass/fail criteria based on requirements, release notes, and known issues.
• Prepare and manage test environments (boards, servers, OS/kernel, software versions) and run tests using provided frameworks and tools.
• Distinguish valid results from interrupted or invalid runs, manage re-tests, coverage, and progress tracking.
• Analyze functional, benchmark, power, temperature and system log data to detect regressions, per-board/server deviations, and abnormal behavior.
• Reproduce failures, isolate whether issues are product defects, environment, or test artifacts, and provide clear reproduction steps and evidence to developers.
• Perform re-verification and regression tests after fixes and document residual impacts.
• Produce evaluation plans, test case documents, execution records, defect analyses, and quality verification reports for release reviews.

Requirements

Must-have technical skills and practical QA experience to plan, run, and interpret system-level tests for NPU software and hardware integration.

• Practical software QA or test design/execution experience; able to convert requirements into verifiable scenarios and pass/fail criteria.
• Ability to define test scope and priorities considering normal/exception/boundary cases and change impact.
• Familiarity with Linux basic commands and using test tools to operate evaluations and inspect logs, processes, and device state.
• Experience using tables/CSV/dashboards to compare results and interpret distributions and basic measurement variance.
• Ability to read and run existing Python or Shell scripts and make simple modifications for test conditions or data processing.
• Understanding of relationships among NPU boards, server environments, firmware, drivers, and SDKs and how these affect test results.
• Clear documentation skills to separate observed facts from hypotheses and to record reproduction steps, expected vs actual results, and analysis rationale.

Nice-to-have (preferred):

• QA experience with AI accelerators (NPU/GPU), SDKs, or hardware-driver integration.
• Experience with LLM serving, vLLM, and performance benchmarking tools.
• Experience in performance regression analysis, per-board variance, power-efficiency, or long-duration stability testing.
• Basic SQL, statistical comparison skills, or experience with Grafana, Superset, Metabase, or similar tools.
• Experience analyzing effects of PCIe, NUMA, temperature, clocking, and throttling on performance.
• Experience improving validation coverage based on customer use cases and operating release verification programs.

Education Requirements

Not specified.

About the Company

Company: FuriosaAI

Headquarters: Seoul, South Korea

FuriosaAI develops high-performance, energy-efficient AI inference hardware and software. Founded in 2017 by semiconductor and AI engineers, the company builds AI-native compute platforms to reduce AI energy and operational costs and operates globally with offices in Korea, Silicon Valley, and an R&D lab in Lisbon.