AI QA Specialist

peopleHum technology Inc.; Bengaluru; On-site; INR 900,000 - 1,300,000; Full time

Job summary

peopleHum technology Inc. seeks an experienced QA Automation Engineer to own end-to-end testing for an AI-native platform.

You will design, implement, and maintain automated tests using Selenium/WebDriver and Playwright, with strong Java coding and API validation across microservices. Focus areas include REST API testing, performance checks with JMeter, and AI testing aspects such as LLMs, guardrails, and AI API validation.

Qualifications

• Hands-on automation ownership using Selenium and Playwright with Java.
• Strong REST API testing with REST Assured, Postman and API validation.
• Experience with performance testing using JMeter and monitoring bottlenecks.
• Proficient in Java; Python/TypeScript a plus for test tooling.
• Familiarity with AI-native platforms, LLMs, guardrail testing and AI API testing.
• Knowledge of Git, microservices, REST/Kafka communication, and SQL basics.

Skills

Selenium WebDriver

Playwright

TestNG/JUnit

REST Assured

Postman

API validation

Apache JMeter

Java

Python

TypeScript/JavaScript

LLMs / AI testing

Git / version control

Microservices / REST-Kafka

SQL basics

Tools

Jenkins

GitLab CI/CD

Docker

Job description

Batch: 2024/2025/2026. Must have skills:

• Strong hands-on experience with Selenium WebDriver, Playwright, and TestNG/JUnit is essential because this role expects end-to-end automation ownership.
• Solid knowledge of REST Assured, Postman, and API validation is a must, since the platform is built on microservices and AI-integrated services.
• Experience with Apache JMeter and understanding response times, throughput, SLA validation, and bottleneck analysis is very important.
• Good coding ability in Java is critical, with added value from Python and TypeScript/JavaScript for automation and test utilities.
• Since this is an AI-native platform, understanding LLMs, AI agents, non-deterministic outputs, guardrail testing, hallucination detection, and AI API testing is one of the biggest differentiators.
• Knowledge of Git, microservices, REST/Kafka communication, and SQL basics is necessary to test reliably in a real engineering environment.

Good to have skills:

• Knowledge of Jenkins and GitLab CI/CD is very valuable because QA automation is much stronger when it is fully tied into build and deployment pipelines.
• Being able to run tests in containers and set up stable test environments is a big plus, especially in modern microservices-based products.
• Since the platform uses event-driven workflows, understanding how to test Kafka message production and consumption is highly useful.
• Experience with Cucumber/Gherkin or Robot Framework helps in writing clear, behavior-driven scenarios that improve collaboration between QA, developers, and product teams.
• Familiarity with Axe-core/WCAG validation and OWASP ZAP adds strong value because it improves both usability and product safety.
• Experience with tools like Claude Code, GitHub Copilot, Cursor, or Cody is a strong advantage for speeding up test automation and working efficiently in an AI-heavy engineering setup.
• Modern code-first performance testing tools are highly valuable because they fit well into CI/CD workflows and are more flexible for engineering teams than traditional load-testing setups.
• This is especially important in microservices-based systems because it helps validate API boundaries and prevents service-to-service integration issues early.
• A very strong plus because it enables isolated, Docker-backed test environments, making integration testing more reliable and production-like.
• A useful frontend-focused E2E testing tool with a strong developer experience, especially valuable for faster debugging and tighter collaboration with frontend teams.
• These are great to have because shift-left security scanning improves product quality early in the pipeline and reduces risk before release.
• This stands out because observability-driven testing is becoming very important in modern distributed systems, especially for tracing failures across services.