AI Software Validation Engineer

As an AI Frameworks Engineer in the Conversion, Performance, and Accuracy team, you will play a critical role in enabling the latest GenAI and Deep Learning models within the OpenVINO™ ecosystem.

• →Model Enabling & Integration: Enable and validate new AI models in OpenVINO, Optimum-Intel, and internal performance and accuracy tools.
• →Benchmarking & Evaluation: Evaluate quantized and native models on Intel hardware, collecting critical metrics for performance, memory consumption, and accuracy.
• →Tool Development: Actively develop and extend opensource benchmarking and validation tools to unblock validation capabilities for new models.
• →Automation & CI/CD Pipelines: Build and maintain dedicated daily pipelines for model enabling, using Jenkins (utilizing Groovy and Python scripting).
• →Automated Log Analysis: Design and implement automated solutions to parse execution logs and pinpoint required changes or additions for successful model enablement.
• →Process & Workload Optimization (Continuous Improvement): Continuously identify bottlenecks in validation workloads, test execution cycles, and infrastructure; drive initiatives to streamline team workflows and operational efficiency.
• →Transparent Reporting & Communication: Deliver clear, actionable status reports, dashboards, and metrics on validation progress, blockers, and model support readiness to team members and project stakeholders.

• Education: Bachelor's degree in Computer Science, Computer Engineering, or a related field with 5–6 years of experience.
• Scripting & Automation: Strong proficiency in scripting languages—primarily Python—and familiarity with Groovy or Bash.
• Problem-Solving & Log Parsing: Strong ability to build regex, parse complex logs, and develop automated scripts to streamline triage and identify execution failures.
• CI/CD Systems: Hands-on experience building or maintaining CI/CD pipelines in Jenkins or similar tools.
• Development Environment: Strong experience with Linux, Git, and Python testing frameworks (e.g., PyTest).
• Model Exploration & Curiosity: Passion for playing with, testing, and adopting new AI models as they are released, with a proven track record of using them in previous work assignments, personal hobby projects, or open-source contributions.

• Inference Toolkits (Great Asset): Familiarity with OpenVINO™, Optimum-Intel, or other deep learning inference runtimes (e.g., ONNX Runtime, TensorRT, llama.cpp).
• Performance & Accuracy Profiling: Experience using benchmarking tools or accuracy check frameworks.
• Model Quantization: Basic understanding of model quantization/compression concepts and evaluating native vs. quantized models.

We offer a total compensation package that ranks among the best in the industry.

*