5+ years as an SDET, QA engineer, or software engineer focused on testing complex distributed systems.
Proven experience in manual and automated testing, including exploratory testing and test planning.
Strong programming skills, preferably in Python, with a focus on clean, maintainable code.
Experience building test frameworks and infrastructure from scratch, not just writing test cases.
Hands-on experience with CI/CD systems and containerized test environments like Docker and Kubernetes.
Experience testing APIs and microservices, including various testing methodologies.
Familiarity with LLM-specific failure modes and understanding of LLM agents' operational mechanics.
Responsibilities
Design and maintain test frameworks and regression suites for LLM-based agents.
Build automated pipelines for continuous evaluation and testing integration.
Develop simulators and sandbox environments for safe agent testing.
Implement various testing methodologies for services and tools that agents rely on.
Conduct hands-on exploratory testing and manual verification of agent behavior.
Benefits
Flexible work environment with opportunities for remote work.
Collaborative culture with a focus on innovation and continuous improvement.
Access to cutting-edge technology and tools in AI and automation.
Opportunities for professional development and mentorship.
Full Job Description
Role Overview
We're looking for a Senior / Staff SDET to own quality for Nexxa's AI agent systems, combining deep test automation with hands-on manual and exploratory testing. Our agents plan, call tools, and take multi-step actions autonomously in industrial environments. You'll write production-grade code to build evaluation harnesses, simulation environments, and CI/CD quality gates, and you'll also get hands-on with the agents yourself: probing behavior, reviewing traces, and validating releases where automation alone can't judge whether the agent did the right thing.
You'll work closely with ML engineers, backend engineers, and Forward Deployed Engineers to define what "good" looks like for an agent operating in high-stakes industrial settings, then build and run the systems and processes that measure it continuously.
Key Responsibilities
Automation and Infrastructure
Design, build, and maintain test frameworks, evaluation harnesses, and regression suites for LLM-based agents, covering reasoning quality, tool-call correctness, task completion, and multi-turn coherence.
Build automated pipelines that run evals and test suites on every model, prompt, or tool-integration change, and integrate them into CI/CD as enforceable quality gates.
Develop simulators, mocks, and sandbox environments for industrial systems and tool APIs so agents can be tested safely, repeatably, and at scale.
Implement API, contract, integration, and end-to-end tests for the services, tools, and orchestration layers that agents depend on.
Develop performance, load, and reliability tests for agent workflows, including failure injection and resilience testing.
Manual and Exploratory Testing
Perform hands-on exploratory testing of agent behavior, including new capabilities, ambiguous or underspecified requests, long multi-turn sessions, and unusual operational scenarios that scripted tests don't anticipate.
Manually review agent traces and outputs to judge reasoning quality, appropriateness of actions, and safety, and use those findings to improve automated checks.
Execute manual regression and release verification passes in staging and customer-like environments before new agent capabilities ship.
Conduct manual red-teaming sessions (prompt injection, jailbreaks, tool misuse, unsafe actions) in partnership with security teams, then convert successful attacks into automated test cases.
Support user acceptance testing and pilot rollouts with Forward Deployed Engineers, capturing real-world failures and feeding them back into test suites.
Write clear test plans, test charters, and bug reports, and maintain manual test documentation where automation isn't yet practical or cost-effective.
Datasets, Metrics, and Quality Strategy
Create and manage golden datasets and labeled test sets, including edge cases, ambiguous inputs, and adversarial prompts specific to industrial and operational contexts, with tooling for versioning and drift detection.
Design rubrics and run human-in-the-loop evaluations where automated scoring isn't reliable enough.
Define and instrument quality metrics beyond simple accuracy (groundedness, hallucination rate, task success rate, latency/cost tradeoffs, and safety violations) and build dashboards to track them over time.
Test agent behavior across the full action loop (planning, tool selection, tool execution, error recovery, and final output), not just the final response.
Investigate and triage failures where the root cause could be the model, the prompt, the tool/API, or the orchestration logic, and turn them into reproducible, actionable bug reports.
Establish quality bars, release criteria, and sign-off processes for new agent capabilities, blending automated gates with manual verification.
Decide what to automate and what to keep manual, and continually shift repeatable manual work into automation.
Mentor engineers on testing strategies and test-driven practices for probabilistic, LLM-driven systems.
Advocate for testability, observability, and tracing in agent architecture from day one.
Qualifications
5+ years as an SDET, QA engineer, or software engineer with a strong testing focus, with demonstrated ownership of test strategy and automation for complex distributed systems.
Proven experience in both manual and automated testing, including exploratory testing, test planning, and writing high-quality bug reports.
Strong programming skills (Python preferred) with a software engineering mindset: clean, maintainable, well-reviewed code for frameworks, data pipelines, and tooling.
Experience building test frameworks and infrastructure from the ground up, not just writing test cases within an existing framework.
Hands-on experience with CI/CD systems (e.g., GitHub Actions, GitLab CI, Jenkins) and containerized test environments (Docker, Kubernetes).
Experience testing APIs and microservices, including contract, integration, and end-to-end testing.
Hands-on experience testing LLM-based products, chatbots, or AI agents, and an understanding of why traditional deterministic assertions break down for generative systems.
Practical experience with eval frameworks or tooling (e.g., promptfoo, DeepEval, RAGAS, LangSmith) or a track record of building your own.
Understanding of how LLM agents work: prompting, tool/function calling, context management, RAG, memory, and orchestration frameworks.
Experience designing test data and labeled datasets, including sourcing, sampling, and managing dataset drift over time.
Familiarity with LLM-specific failure modes: hallucination, prompt injection, context poisoning, tool misuse, goal drift, and non-determinism.
Comfort operating in ambiguity, defining what "correct" means for a task when there's no single right answer.
Strong written communication skills for turning fuzzy quality signals into clear, actionable findings for engineering and product stakeholders.
Experience with performance and load testing tools (e.g., k6, Locust) and chaos or fault-injection testing.
Background in ML/data science sufficient to read model evals and statistical significance.
Experience red-teaming or doing adversarial/security testing on ML systems.
Familiarity with observability and tracing tools for LLM applications (e.g., LangSmith, Arize, Langfuse, Weights & Biases) and general observability stacks (OpenTelemetry, Grafana, Datadog).
Experience testing systems in industrial, IoT, or operational technology (OT) environments, including industrial protocols (OPC UA, MQTT, Modbus) or simulation and hardware-in-the-loop setups.
Experience supporting UAT, pilots, or customer-facing validation of complex products.
Experience setting up test and eval infrastructure from scratch at a startup or fast-moving team.
What We're Looking For
An engineer who is equally comfortable writing a test framework and sitting down to break an agent by hand.
Someone who can build rigorous evaluation systems for problems that don't have a single right answer, and who knows when human judgment is the right tool.
A systems thinker who enjoys turning ambiguous agent behavior into measurable, trustworthy signals.
A strong collaborator who partners well with ML engineers, backend engineers, and Forward Deployed teams.
