Sr Software Engineer

SDET / Agentic Test Automation Engineer
Experience: 5-8 years
Location: Bengaluru, India
Employment: Full-time
Focus: AI/Agentic Systems - Test Automation
Quality Engineering - Python - Distributed Systems
About the Role
We are hiring an SDET / Agentic Test Automation Engineer to establish the quality engineering foundation for
an AI and agentic software platform. This is an engineering role centred on building test systems and
infrastructure, rather than on test execution or script maintenance.
You will be responsible for validating a platform composed of AI agents, agentic workflows, LLM integrations,
RAG pipelines, distributed services and asynchronous infrastructure. The work involves determining how these
systems behave under concurrency, at scale, under adversarial input and during partial failure -- and building
the automation required to verify that behaviour continuously.
Responsibilities
- Architect and build scalable test automation frameworks in Python, including the supporting libraries,
fixtures, simulators and test infrastructure.
- Own API, integration, end-to-end, component and system-level testing strategy across the platform.
- Deliver deterministic, reproducible and debuggable tests, and maintain environments that allow engineers to
reproduce failures quickly. Diagnose and remove flaky tests rather than tolerate them.
- Design testing approaches for AI and agentic systems, covering LLM-powered agents, multi-agent
workflows, tool calling, RAG pipelines, memory and context systems, model routing, prompt and version
regressions, structured outputs, long-running workflows, human-in-the-loop flows, and model and tool failure
fallbacks. Tests must assess whether outcomes are correct, safe, consistent and useful -- not only whether
execution completes.
- Apply and build advanced testing techniques to surface defect classes that conventional test cases miss:
property-based, fuzz, metamorphic and differential testing, golden datasets, contract and state-machine
testing, fault injection, chaos testing, and concurrency and race-condition testing.
- Test distributed system failure modes as well as components -- retries and idempotency, timeouts, partial
failures, service dependencies and data consistency across queues, workers, caches and databases.
- Build automated performance, load, stress, soak and scalability testing to establish capacity and failure
limits ahead of production exposure.
- Integrate testing into CI/CD so that quality gates are enforced continuously.
Page 1 of 2
Confidential -- For Internal & External Recruitment Use Only
SDET / Agentic Test Automation Engineer -- Job Description
- Partner with backend, frontend, AI/ML, platform and infrastructure teams to make quality measurable,
observable and enforceable.
Required Skills
- 5-10 years in an SDET, software engineering or automation-focused engineering role.
- Expert-level Python, or strong proficiency in C++, Java, Go or Rust with willingness to work extensively in
Python.
- Demonstrated experience architecting automation frameworks, not solely writing tests within an existing
one.
- Strong fundamentals: data structures and algorithms, object-oriented design, concurrency and
multithreading/multiprocessing, networking, HTTP/REST/WebSockets, SQL and databases, Linux and Git.
- Experience testing distributed or cloud-native systems, including microservices, message queues,
asynchronous pipelines, caches and databases.
- Strong command of pytest or equivalent frameworks, API and integration testing, and CI/CD integration.
- Proven debugging and root-cause analysis across application, database, network and infrastructure layers.
Preferred Skills
- Kubernetes and Docker; Azure, AWS or GCP.
- FastAPI or comparable backend frameworks; PostgreSQL; Redis; Kafka or other messaging systems.
- Load and performance tooling such as Locust, k6 or JMeter; browser automation with Playwright or
Selenium.
- OpenTelemetry; GitHub Actions or Jenkins.
- Property-based testing with Hypothesis.
- AI/ML testing and evaluation, LLM APIs, RAG and vector databases, and agent frameworks such as
LangChain or LangGraph.
- Chaos engineering, security testing, or experience building testing infrastructure from the ground up