Headline

AI / LLM QA Engineer | Hallucination Detection & RAG Testing | Fintech QA | Prompt & Adversarial Testing | BINOCS

About

AI / LLM QA Engineer with 1+ year of experience owning end-to-end quality on a live fintech AI platform at BINOCS. Promoted from QA Intern to QA Engineer based on performance. I specialise in testing LLM-powered workflows — hallucination detection, RAG output verification, adversarial prompt testing, and golden dataset comparison. Alongside AI testing, I handle REST API validation, defect lifecycle management, and full functional QA across Agile sprint cycles. Key highlights: • Caught 15+ high-severity LLM hallucination bugs before production • Blocked 8+ critical API contract violations before UAT • Designed 650+ test cases adopted as team-wide QA templates • Promoted Intern → Engineer within months based on independent performance Currently expanding expertise in LLM evaluation frameworks and test automation.

Location

Bengaluru

AI stack

claude

cursor

openai

Work experience

QA Engineer — AI / LLM Testing at Binocs.co, Aug 2025 - Present · 1 yr 3 mos, AI & LLM Validation: • Spearheaded end-to-end testing for AI-driven financial reporting modules and conversational AI workflows. • Validated AI responses by identifying and documenting prompt hallucinations and refining AI-generated outputs for accuracy. API Validation & Security Testing: • Leveraged AI tools (Claude/ChatGPT) to generate security-focused test cases, identifying critical vulnerabilities such as SQL injection risks in payment gateway APIs. • Prevented potential internal data exposure by identifying unhandled API error responses that leaked backend Go implementation details. • Utilized Postman for deep REST API validation, testing payloads, status codes, and executing data-driven workflows to ensure backend data integrity. • Tested browser console manipulation to ensure transactions were validated strictly on the backend, preventing simulated fake payment flows. Fintech Workflows & Core QA: • Designed, reviewed, and executed 500+ manual test cases across fintech modules, including highly parameterized test cases for complex Stripe payment integrations. • Managed the complete defect lifecycle (500+ defects) in JIRA, collaborating directly with developers to resolve backend workflow and complex financial calculation errors. • Performed regression, smoke, sanity, and exploratory testing across Agile sprint cycles

Quality Assurance Intern at Binocs.co, Feb 2025 - Aug 2025 · 7 mos

Education

Bachelor of Technology - BTech, Mechatronics, Jain (Deemed-to-be University)

Polytechnic, Mechanical Engineering, Government polytechnic anantapur