QA Engineer (AI Products)

QA Engineer (AI Products) — Kickker AI

Location: India (Gurgaon) | Team: Tech | Reports to: CTO

About Kickker AI

We're building the first global AI platform at the intersection of conversational AI, browser automation, and live video/voice calls. Our AI agents autonomously navigate and control live workflows, turning customer conversations into completed demos, onboarding, and checkout experiences. Founded by ex-FAANG & Naukri.com alumni with 25+ years of building enterprise-grade production systems.

Live with initial customers and growing fast, we've strong early traction within 45 days of launch.

The Role

Testing AI is not like testing traditional software. Outputs are non-deterministic, edge cases are infinite, and "correct" is often a spectrum. As a QA Engineer at Kickker AI, you'll own the quality of AI agents that interact with real customers in real-time — over live video calls, browser automation, and voice. You'll need to think beyond test cases and build systems that catch failures before customers do.

You'll be one of the founding QA engineers — defining what quality means for an AI-native product from the ground up.

What You'll Do

• Test AI agent rollouts end-to-end — from knowledge ingestion and conversation quality to browser navigation accuracy and live video interactions
• Build evaluation frameworks — define what "good" looks like for non-deterministic AI outputs (conversation quality, navigation accuracy, response relevance, hallucination detection)
• Design and execute test strategies for multi-modal systems — voice, video, browser automation, and their real-time interplay
• Catch what's hard to catch — identify hallucinations, incorrect browser actions, conversation dead-ends, latency spikes, and edge-case failures
• Automate ruthlessly — build automated test suites for regression, performance, and integration testing across the agent pipeline
• Test in production context — validate agent behavior across different client deployments, knowledge bases, and use cases
• Collaborate with AI engineers and FDEs — reproduce client-reported issues, triage bugs, and ensure fixes actually fix things
• Own release quality — every agent that goes live passes through you

What We're Looking For

• 1-5 years of QA experience, with meaningful exposure to AI/ML product testing (junior and mid-level both welcome)
• You understand that testing AI is different — you've dealt with non-deterministic outputs, model evaluation, or prompt testing
• Experience with test automation frameworks (Selenium, Playwright, Pytest, or similar)
• Comfort with Python — you can write scripts, parse logs, and build test utilities
• Exposure to any of: LLM testing, chatbot QA, browser automation testing, or voice/video product testing is a strong plus
• Strong attention to detail combined with the ability to prioritize what matters — you can't test everything, so you test the right things
• You're opinionated about quality but pragmatic about shipping

This Role is NOT For You If

• You only know manual testing with static expected outputs
• You're uncomfortable with ambiguity — AI products don't always have a single "right answer"
• You see QA as a checkbox before release, not as a continuous system

Why Join Now

• Define QA for AI agents — there's no playbook; you'll write it
• High-stakes testing — your work directly protects live customer interactions; when an agent fumbles a demo on a video call, it's visible and immediate
• Unique testing challenges — multi-modal, real-time, non-deterministic systems that combine browser automation + voice + video — few QA engineers in the world are solving this
• Early team — shape quality standards and testing culture as one of the first QA engineers
• Founder access — work directly with engineers who've built production AI at Google, Microsoft, Amazon and Infoedge.