Blue Machines

- Full-time

About Blue Machines

Blue Machines builds the platform for production AI agents — agent builder, real-time voice agents, workflow orchestration, integrations, and inference. Multi-tenant systems, non-deterministic behaviour, many providers in the call path, and high bar for what ships to customers.

About The Role

We are hiring a founding Quality Engineering Lead — a builder who designs and ships the systems that decide what is safe to release. Your output is platforms and mechanisms, not test cases. You may grow a team later; you are hired to build first.

What You Will Do

- Design and build the test platform: execution, isolation, test-data lifecycle, reporting, flake handling, and a clear answer to whether a build should ship

- Build an agent evaluation harness that detects behavioural regressions after prompt, model, tool, or config changes

- Own release gating: enforceable gates in code, post-deploy verification, canary criteria, and a narrow hotfix path

- Treat capacity and load as release criteria, not optional reports

- Turn customer-visible defects into durable detection (owned, staged, measurable)

- Push testability into product systems — seams, replay, fault injection, observability





- Raise the engineering bar through review and mechanisms, not by writing every test yourself

What we are looking for

- Robust software engineering background — you can design a system from an ambiguous problem and implement the hard path

- Experience owning quality, reliability, release, or evaluation platforms (not only writing suites)

- Experience with conversational AI, voice, or LLM evaluation — or a clear track record evaluating non-deterministic systems without ground truth

- Release engineering, progressive delivery, or SRE depth — gates, canaries, and production signal you have owned

- Hands-on building of CI/CD quality infrastructure at meaningful scale

- Product mindset — you tie quality work to customer-visible risk and delivery outcomes

- Evidence over assertion — you quantify, and you separate known from assumed

- Ability to work with a small team and influence engineers who do not report to you

This role is not A manual QA role, a checklist sign-off role, or a hire to write Playwright/pytest cases as the main job. We already have deep automation capability; we need the architect who builds what sits above it.

📌 Quality Engineering Lead (Bengaluru)
🏢 Blue Machines AI
📍 Bengaluru