AI System Quality Assurance Engineer – Data & Agentic AI

Role– AI System Quality Assurance Engineer – Data & Agentic AI Experience: 1–3 years Positions: 2 Role Summary

• This role will focus on validating data movement, mapping, transformation, and integrity across migration testing and the integration between our agentic AI quoting platform and Submission. Link. AI agents consume Small Business Owner information provided through Submission. Link and backed by structured data models to create insurance applications. The role will verify what data is being used, how it moves through the system, and whether it is correctly validated at each stage, from source payloads and migration outputs through APIs, AI agents, business rules, guardrails, and the user interface. It requires strong API, integration, and data-validation skills, along with practical exposure to agentic testing and validation of AI-generated outputs. The candidate must have strong communication skills and be capable of working directly with US-based Product, Data Intelligence, Engineering, and business stakeholders.

Key Responsibilities

• Understand what data is being used and map how it moves through each stage of the system pipeline Perform data validation during migration testing, including source-to-target comparison, completeness checks, accuracy checks, and transformation validation Validate Submission. Link data against expected Small Business Owner information and downstream insurance-application outputs Verify that AI agents correctly consume, interpret, and apply Submission. Link data when creating insurance applications Validate data mapping and transformation across source systems, APIs, AI agents, business rules, guardrails, and the UI Build or execute validation scripts to check data integrity, completeness, schema conformance, and transformation accuracy across the pipeline Test agentic workflows and validate AI-agent decisions, tool usage, fallback behavior, exception handling, and human-review handoffs Define and execute validations for AI-generated outputs, including checks for missing, incorrect, inconsistent, unsupported, fabricated, or policy-violating information Validate AI outputs against defined business rules, data models, guardrails, expected outcome ranges, and acceptance thresholds Determine whether issues originate in source data, migration logic, data models, integration layers, AI-agent behavior, guardrails, or front-end presentation Develop integration and regression tests covering common, negative, edge-case, and AI-output validation scenarios Work directly with US-based Product, Data Intelligence, Engineering, and business teams Clearly communicate defects, evidence, quality risks, guardrail gaps, and test findings to stakeholders Requirements Required Experience and Skills 1–3 years of QA experience, with a strong focus on API, integration, data validation, or migration testing Proven experience validating data integrity, completeness, accuracy, and transformation during migration testing Ability to understand what data is being used, trace it from source to target, and validate it as it moves through system workflows Strong experience validating complex data models, API payloads, JSON structures, mappings, and schema transformations across multiple systems Experience testing AI, LLM, or agentic AI applications, including non-deterministic and rules-driven outcomes Exposure to agentic testing approaches, including validating AI-agent decisions, tool usage, fallback behavior, and workflow outcomes Experience defining or validating guardrails, acceptance criteria, and validation checks for AI-generated outputs Strong API testing experience using tools such as Postman Ability to create detailed test cases focused on data integrity, mapping, transformation, AI-output validation, and edge-case scenarios Strong functional, integration, regression, analytical, investigative, and defect-isolation skills Strong verbal and written communication skills, with the ability to work directly and independently with US-based stakeholders Ability to explain data-integrity issues, AI behavior, guardrail failures, and non-deterministic outcomes to technical and non-technical stakeholders Preferred Experience Insurance domain experience, preferably in commercial insurance, quoting, underwriting, or insurance application workflows Familiarity with Model Context Protocol (MCP) Experience validating structured data consumed or generated by AI systems Understanding of AI evaluation methods, acceptable outcome ranges, hallucination checks, validation thresholds, and guardrail effectiveness SQL or similar data-querying skills