AI Full Stack Manual QA Engineer

Qapitol QAS Private Limited; Bengaluru; On-site; INR 700,000 - 1,100,000; Full time

Job summary

Qapitol QAS Private Limited in Bengaluru is seeking a QA professional to own end-to-end manual testing across web applications, APIs, backend services and databases.

You will validate AI-enabled features, spot hallucinations and prompt/response mismatches, and log defects with clear reproduction steps, triaging them through to resolution.

The role emphasizes systems thinking, tracing data flows from UI to API to database, ensuring quality before releases in a fast-paced tech environment.

Qualifications

• Willing to own end-to-end manual testing from requirements to sign-off.
• Ability to trace defects across UI, API, SQL, and AI features.
• Curious about AI/GenAI concepts and how they affect test outcomes.
• Excellent problem-solving and edge-case detection.

Responsibilities

• Conduct end-to-end testing across web applications, APIs, backend services and databases.
• Create and maintain signed-off test cases and log defects.
• Validate AI-enabled features for relevance, correctness, and consistency.
• Identify hallucinations and prompt/response mismatches.
• Collaborate with developers and product owners to triage issues.

Skills

End-to-end testing

Manual testing

AI/GenAI concepts

Defect logging & triage

Systems thinking

Job description

• Website team: Professional Services BU
• Location on website: Bengaluru · onsite
• Suggested Keka location: Bengaluru (Onsite)
• Job type: Full Time
• Experience: 3 to 5 years

You test the whole system end to end - web, API, database and the AI features sitting on top of it - and you catch what breaks before a client does.

Most QA hires can test a UI or run a script. This role is for someone who can trace a defect through every layer it touches: a UI break in the browser, a request/response mismatch in the API layer, a bad value in a SQL query, and now, increasingly, an AI feature returning an answer that isn't quite right. AI-enabled applications don't remove the need for rigorous manual testing - they add a layer of validation on top of it, and this role sits at the meeting point of the two.

You will own end-to-end manual testing across web applications, APIs, backend services and databases, working from business requirements through to signed-off test cases and defects logged, triaged and retested. Alongside that, you will bring a working understanding of AI/GenAI concepts to validate relevance, correctness, consistency and expected behaviour in AI-enabled features - spotting hallucinations, prompt/response mismatches and inconsistent outputs the way you'd spot any other defect.

Who thrives here

Testers who think in systems, not screens - who instinctively ask what happens to a value as it moves from UI to API to database and back. You're comfortable being the person who finds the edge case nobody planned for, and curious enough about AI to want to understand how a feature is supposed to behave before deciding it's wrong.