CEHPOINT
Book a consultation

Full-time · Remote (India)

GPU Infrastructure & MLOps Engineer

Cehpoint is hiring for this role: full-time, Remote (India), ₹12,00,000 – ₹21,00,000 / year. Run the GPUs and serving stack behind Cehpoint AI, and help plan the compute campus in West Bengal.

Apply for this roleCheck your skill fit

About the role

Cehpoint AI serves chat, agents, images and video through our gateway. You will run inference reliably and cheaply today, and help design the compute and cooling plan for the sustainable compute campus we are studying in West Bengal.

Run the GPUs and serving stack behind Cehpoint AI, and help plan the compute campus in West Bengal.

What you will do

  • Deploy, scale and monitor model-serving workloads on GPU servers.
  • Measure cost per request, latency and failure rates, and bring them down.
  • Automate deployment, backups and alerting so the platform keeps running when someone is asleep.
  • Contribute to sizing studies for power, cooling and GPU capacity for our planned facility.

What you should know

  • Linux administration, containers and at least one way of serving models in production.
  • How GPU memory, batching and quantisation affect throughput.

What you should be able to do

  • Debug a production incident from logs and metrics and write a clear post-mortem.

How you work

  • You care about watts and rupees per request, not only uptime.

Skills we check

  • Linux
  • Docker
  • Kubernetes or Nomad
  • NVIDIA drivers and CUDA
  • Model serving (vLLM, TGI or similar)
  • Monitoring
  • Networking

What you get

  • Hands-on ownership of production AI infrastructure.
  • Early involvement in a data centre design study.

How we select for this role

We read every application ourselves. Here is what happens after you apply.

  1. Apply with a short form

    Paste a link to your CV and answer five questions in your own words. No sign-in and no file upload. About fifteen minutes.

  2. We read your application

    A person on the team reads your answers and your CV, usually within a few working days.

  3. Conversation with the team

    If your profile fits, we email you to set up a video conversation about your work and this role.

  4. Practical project

    Where the role needs it, a practical project of up to 5 working days, reviewed by an engineer. For this role: serve a supplied open model behind an OpenAI-compatible endpoint, load-test it and report cost per thousand requests.

  5. Offer

    If it is a match on both sides, you receive your offer letter by email and in the portal.

Other open roles in this team

See all open roles