Manual Tester - QA

We are seeking a high-agency, detail-oriented Manual QA Engineer to join our core AI evaluation team at Scaler AI Labs. In this role, you will be responsible for validating next-generation AI platforms, evaluating Reinforcement Learning (RL) environment tasks, and rigorously testing automated pass/fail grader logic to prevent false positives and model hallucinations. You will collaborate closely with AI researchers, product managers, and software engineers to ensure high standards of quality, reliability, and safety across web, mobile, and API-driven AI workflows.

Responsibilities:

• AI Evaluation and Grader Logic QA: Evaluate task prompts for clarity and ambiguity, and design rigorous test suites to test automated pass/fail grader scripts in RL environments.
• Adversarial and Edge-Case Testing: Identify model failure modes, edge cases, and hallucinations; verify that automated graders correctly reject incomplete or malformed outputs (mitigating false passes).
• API and Backend Validation: Perform hands-on REST API validation using Postman/Swagger, inspecting endpoints, headers, JSON payloads, schema structures, and HTTP status codes.
• Database and Integrity Verification: Write and execute SQL queries to validate data pipelines, model inputs/outputs, backend databases, and data consistency.
• Issue Debugging and Root Cause Analysis: Utilize log debugging tools (Charles Proxy, Android Studio, DevTools, and system logs) to isolate defects between the frontend UI, API contracts, and backend microservices.
• Defect Management and STLC Ownership: Own test scenario design, test case authoring, traceability matrices (RTM), and end-to-end defect tracking to closure in JIRA within an Agile/Scrum environment.

Requirements:

• Experience: 2 to 5 years of solid hands-on experience in Manual, Functional, and System QA.
• API Testing: Proficient in API testing with Postman, including assertions, environment variables, token authentication, and payload contract validation.
• Database Testing: Strong practical knowledge of SQL (SELECT queries, JOINS, data filtering, and backend verification).
• Debugging Tools: Hands-on exposure to network proxy and device debugging tools such as Charles Proxy, Android Studio logcat, or browser DevTools.
• AI / LLM Acumen: Familiarity with LLM interaction, prompt engineering concepts, or testing conversational AI/AI agent integrations.
• Work Model Commitment: Willingness and availability to work on a 6-day schedule (5 days working from the Jayanagar office + 1 day WFH).
• Availability: Immediate joiners or candidates with a notice period of 15 days or less are preferred.