How AI Agents Test Apps Autonomously, Without Scripts
ContextQA (28+ yrs in QA)NKNaveen Khunteta
Host, Naveen AutomationLabs
Autonomous AI agents can now test applications from plain-English intent instead of predefined scripts. In this live session with Naveen AutomationLabs, ContextQA's Mukund Wangikar demos agentic testing, knowledge-graph-driven generation, and shift-right AI Insights — while tackling the community's biggest fear: will AI replace testers?
Walk away knowing how to apply it
What the conversation covers
Autonomous/agentic testing: an executable test from a 3-word English prompt
"QA isn't dying, it's evolving" — the human-machine intelligence gap
Stopping hallucination with LLMs trained on testing data and human-in-the-loop
Multi-source generation: BRD/SRS, Jira stories, and an uploaded video walkthrough
Knowledge graphs that update, add, or deprecate tests when requirements change
ContextQA's specialized agents for generation, automation, and root-cause analysis
Code export to popular frameworks, self-healing scores, and parallel execution
AI Insights (shift-right): telemetry-only data that surfaces uncovered user flows
The ideas worth remembering
The tester's role shifts from executor to strategist and orchestrator
Test-purpose-trained models and well-built agents are what stop hallucination
AI matches development speed for regression; exploratory testing stays human
Context matters most — feed agents requirements, stories, even walkthroughs
AI will not take your job. The person who knows AI will take your job.— Mukund Wangikar
Who you'll hear from
Mukund Wangikar
ContextQA (28+ yrs in QA)
Naveen Khunteta
Host, Naveen AutomationLabs
See ContextQA in action
Go from watching to doing — spin up an AI agent and watch it test, self-heal, and report for you.