AI Software Validation Engineer

The Software Group at Intel develops tools and platforms focused on AI inference on Intel hardware such as CPUs, GPUs, and NPUs. The team ensures compatibility and performance of the OpenVINO™ ecosystem with state-of-the-art AI models.

• Bachelor's degree in Computer Science, Computer Engineering, or related field with 5–6 years of experience.
• Strong proficiency in scripting languages, primarily Python, with familiarity in Groovy or Bash.
• Ability to build regex patterns and parse complex logs; develop automated scripts for triage and failure identification.
• Hands-on experience with CI/CD pipelines using Jenkins or similar tools.
• Experience with Linux, Git, and Python testing frameworks such as PyTest.
• Passion for exploring, testing, and adopting new AI models.

Preferred Qualifications:

• Familiarity with inference toolkits such as OpenVINO™, Optimum-Intel, ONNX Runtime, TensorRT, or llama.cpp.
• Experience in benchmarking and accuracy profiling tools.
• Basic understanding of model quantization and evaluation of native vs. quantized models.

• Enable and validate new AI models in OpenVINO, Optimum-Intel, and internal tools.
• Evaluate quantized and native models on Intel hardware, collecting metrics on performance, memory consumption, and accuracy.
• Develop and extend open source benchmarking and validation tools.
• Build and maintain daily automation pipelines for model enabling using Jenkins with Groovy and Python scripting.
• Design and implement automated log parsing solutions to identify necessary changes for successful model enablement.
• Identify bottlenecks in validation workloads and test cycles, driving process and operational improvements.
• Deliver clear status reports, dashboards, and metrics on validation progress and model readiness to stakeholders.

All qualified applicants will receive consideration for employment without regard to protected characteristics as per local law and regulations. The role follows Intel's hybrid work model, allowing employees to split time between on-site and off-site work. Job posting details such as work model, location, or time type are subject to change.