Senior Software Development Test Engineer - AI

About the Role
We are looking for highly motivated Senior AI Software Development Engineers in Test (AISDET) to join Tekion’s ML Engineering team. Our team builds and ships the machine learning 
models and AI agents that power the Automotive Retail Cloud (ARC) — from Service AI Agents and F&I automation to the AI-powered Analytics Agent — trained on real-time data spanning 1,700+ dealership rooftops.
In this role, you will be responsible for building confidence in the quality, correctness, and reliability of the models and AI agents we deliver. You will work closely with ML Engineers, Data
Scientists, the AI Platform team, Product Management, and Applied AI teams to define robust validation strategies, automate testing across the model lifecycle, and ensure our AI features meet the highest quality bar before they reach dealers and their customers.Because you are testing AI systems, you will go beyond deterministic assertions: you will design
evaluation frameworks for non-deterministic model and agent outputs, measure accuracy and consistency, catch hallucinations and regressions, and continuously expand coverage. You will also use AI and LLMs yourself — to generate test cases, synthetic data, and automation scaffolding — raising both the speed and the depth of quality engineering.
What You’ll Do

• Develop a deep understanding of Tekion’s AI agents, ML models, business domains, and the data that feeds them.
• Own end-to-end quality for the ML models, AI agents, and AI-powered features shipped by the ML Engineering team.
• Design and implement automated testing frameworks for model inference, prompt pipelines, agentic workflows, RAG/retrieval systems, and the APIs that serve them.
• Validate model correctness, accuracy, latency, and behavioral consistency across model versions, prompt changes, and upgrades.
• Define and execute evaluation strategies for AI/LLM-powered capabilities by:
• Creating and curating evaluation datasets (evals) and ground-truth sets
• Measuring response accuracy, relevance, and consistency
• Identifying hallucinations, unsafe outputs, and edge cases
• Building LLM-as-judge and automated scoring pipelines
• Continuously improving evaluation coverage as models and agents evolve
• Use AI/LLMs to accelerate quality engineering — generating test cases, synthetic test data,and automation scaffolding.
• Build regression suites that detect quality drift when models, prompts, embeddings, or training data change.
• Validate integration of ML models and agents into ARC products, including end-to-end flows and fallback behavior.
• Drive root cause analysis for production model and behavioral issues, and partner with ML Engineers on preventive improvements.
• Define quality metrics, guardrails, and automated monitoring to detect model degradation before it impacts customers.
• Champion quality and evaluation engineering best practices across the ML Engineering organization.

What We’re Looking For

• 5–8 years in SDET, quality engineering, or software engineering, with a strong track recordof building test automation frameworks.
• Strong programming skills in Python (preferred for ML tooling) and/or Java, with the ability to write production-quality automation code.
• Hands-on experience testing data-intensive or ML/AI systems — or a strong softwaretesting background with demonstrated ML/LLM fluency.
• Solid understanding of ML concepts: model training and inference, evaluation metrics (precision/recall, F1, etc.), and the non-deterministic nature of model outputs.
• Experience designing evaluation frameworks, or working with eval datasets, benchmarks, or LLM-as-judge approaches.
• Familiarity with LLM/agent concepts — prompting, RAG, embeddings, vector search — and generative failure modes (hallucination, drift, prompt sensitivity).
• Experience with CI/CD, test orchestration, and API/integration testing.
• Strong analytical and debugging skills; comfortable performing root cause analysis across model, data, and code layers.
• Excellent collaboration and communication skills to partner across ML, data, and product teams.

Nice to Have

• Experience with ML/eval tooling such as MLflow, Weights & Biases, LangSmith, Ragas, DeepEval, or provider eval suites.
• Experience with cloud platforms (AWS/GCP/Azure) and containerized environments (Docker, Kubernetes).
• Exposure to model observability, drift detection, and production ML monitoring.
• Experience testing agentic systems, tool use, or multi-step LLM workflows.
• Prior experience in a high-scale SaaS or data platform environment.