Nitish Srinivasan

Quality Engineering Analyst with 3.9 years of experience testing and validating Generative AI applications, Large Language Models (LLMs), RAG pipelines, vector retrieval systems, and AI agent workflows on enterprise platforms. Skilled in executing AI-generated test cases, validating AI-assisted test tooling outputs, prompt testing, hallucination detection, conversational AI validation, and structured LLM output evaluation, while documenting findings and building reusable AI testing assets for the broader test team. Proficient in REST API testing (Postman), manual test execution, defect lifecycle management, and test strategy & planning using Azure DevOps and Jira.

Comfortable working within Agile/Scrum delivery teams and collaborating with onshore stakeholders and client teams; a quick learner who adapts fast to new enterprise test platforms and tooling. Known for strong analytical thinking, attention to detail, and cross-functional collaboration in delivering high-quality QA outcomes across GenAI and non-AI projects.

Quality Engineering Analyst at Accenture (2022-11 – Present)

Project: GenWizard – Gen AI + MyWizard — a Generative AI-powered platform integrating project management tools to streamline and automate SDLC workflows across enterprise delivery.

• Executed AI-generated test cases and validated AI-assisted test tooling outputs across LLMs, RAG pipelines, vector-based retrieval systems, and AI agent workflows within a GenAI-powered project management integration platform.
• Applied structured model evaluation criteria — relevance, contextual grounding, hallucination detection, and response consistency — to validate LLM outputs, identifying 45+ AI-related defects across multiple releases.
• Validated RAG pipelines by testing embedding-based document retrieval, semantic relevance, and ranking accuracy, reporting 30+ retrieval and grounding issues across multiple pipeline configurations.
• Identified 20+ integration and logic defects across AI agent workflows by evaluating task execution, dependency resolution, workflow sequencing, and failure-recovery scenarios.
• Conducted end-to-end validation of AI-generated SDLC artifacts, including requirements, design documents, architecture outputs, test cases, and code suggestions, ensuring correctness, traceability, and consistency across project releases.
• Verified application functionality after code migrations to higher versions, confirming output consistency and optimizing prompts to maintain expected results.
• Assisted in conversational AI testing across 100+ multi-turn test scenarios, validating context retention, response accuracy, and user intent alignment across varied input conditions.
• Documented findings and built reusable, modular AI testing assets and test scenarios for LLM and non-AI modules, reducing test cycle time and improving release velocity for the broader test team.
• Carried out functional, integration, API, smoke, sanity, and regression testing across AI-driven and non-AI modules; executed REST API testing using Postman to validate response behavior and data integrity.
• Logged, tracked, and retested defects using Azure DevOps, ensuring accurate severity classification, defect lifecycle management, and closure validation aligned with project delivery timelines.
• Collaborated cross-functionally with developers and onshore stakeholders to troubleshoot LLM defects and refine prompts, contributing to improved AI response quality and reduced hallucination rates across releases.
• Applied basic Selenium scripting to support automation of repetitive test scenarios, complementing manual QA efforts.
• Contributed to test strategy and planning by documenting detailed test cases and scenarios, supporting release readiness and UAT sign-off.

Bachelor of Science in Computer Science and Applications – Sri Krishna College of Arts and Science (2019-06 – 2022-06)