LLMOps Engineer

As an LLMOps Engineer at Thinqmesh, you will keep AI systems reliable once real people depend on them. You will build the platform that deploys, traces, evaluates and controls the cost of every model-powered feature we run.

Your work makes it safe for the whole team to move quickly, because every change is tested and every answer can be traced.

As AI moves from pilots into production, the hard problems become operational: models change, costs climb and failures are subtle. Observability and cost optimisation now appear as core requirements in AI job postings, and organisations need engineers who can apply the discipline of DevOps to systems that behave differently from ordinary software.

You will build deployment and release pipelines for AI applications, including versioning of prompts and models. You will run evaluation in continuous integration, set up tracing and alerting, and track quality, latency and cost in production.

You will design for failure with fallbacks and rate limits, manage model hosting where it is needed, and put in place the controls that keep data and access secure.

You bring at least three years in DevOps, platform or machine learning engineering, with strong cloud, container and CI/CD experience, and practical knowledge of how LLM systems behave in production. You write good Python and you enjoy making complex systems dependable.

Experience with model serving, GPU infrastructure, or tools such as Langfuse or Braintrust would be a clear advantage.

This role is hybrid or remote. It is full time in Noida, and a consultant role in San Francisco, Tustin and London. You will work closely with the client teams and colleagues you support.

Production systems sometimes need attention outside normal hours, and we share that fairly across the team.

We read every application ourselves. Show us something you have built or shipped, because real work gives us far more to talk about than a list of tools.