Our Client, a Health Insurance company, is looking for a Software Engineer for their Columbia, MD/Hybrid location.
Responsibilities:
Conduct deep technical analysis and testing of AI/ML and Generative AI systems, including validation and verification of model outputs, prompt behavior, and end-to-end application workflows.
Participate in Business Requirements reviews and collaborate with systems analysts and developers to fully test application changes, including AI/ML model integrations and Gen AI features.
Understand functional specifications and scope of releases to design robust testing harnesses that ensure quality delivery of AI-powered applications.
Write complex SQL queries for retrieval of data from multi-database environments, including querying ML feature stores and model metadata. (Strong knowledge of Oracle database required).
Apply expert systems analysis and design techniques to complex enterprise systems, with a focus on validating Gen AI outputs, LLM responses, and ML model predictions.
Maintain broad knowledge of data sources/flow, interactions of complex systems, and the capabilities and limitations of AI/ML systems, including model drift, hallucination detection, and bias evaluation.
Work closely with Product Owners, developers, Data Scientists, and stakeholders to understand AI/ML product specifics and create corresponding test strategies, including evaluation frameworks for generative outputs.
Execute manual and exploratory testing on frontend (UI) and backend (API) systems, including AI/ML inference endpoints and Gen AI prompt pipelines.
Develop and maintain testing suites for a large-scale Salesforce environment integrated with AI/ML capabilities such as Salesforce Einstein and external LLM services.
Responsible for front-end and back-end testing for applications built on the Salesforce platform, including AI-driven features and automated decision systems.
Requirements:
Detailed knowledge of Salesforce applications, ML model lifecycle, and Generative AI concepts including LLMs, prompt engineering, RAG (Retrieval-Augmented Generation), and vector databases.
3+ years of relevant testing experience with a strong background in Software QA with cloud technologies, including 3+ years of experience with hands-on exposure to AI/ML application testing.
2+ years of relational database experience including generating complex queries, setting up database connections, and automating query processing, with familiarity in querying ML feature stores or data pipelines.
Experience in automated and manual testing in cloud environments, particularly for AI/ML and Gen AI applications, including validation of model outputs, embeddings, and LLM-generated responses.
Experience designing test strategies for non-deterministic systems, including techniques for evaluating accuracy, relevance, and safety of Gen AI outputs.
1+ years of experience with BDD/Selenium and automation tools, with a plus for experience in AI testing frameworks.
Experience using Agile team collaboration and requirements management tools (Jira, Confluence).
Excellent written and oral communication skills and problem-solving abilities, with the ability to clearly articulate AI/ML testing findings to both technical and non-technical stakeholders.
Strong aptitude to learn and adapt to emerging AI/ML technologies and evolving Gen AI tooling.
Familiarity with AWS AI/ML services (SageMaker, Bedrock) or equivalent cloud AI platforms is a plus.