Full-time · Remote (India)
Cognitive Infrastructure Engineer
Optimize high-throughput LLM serving infrastructure, tensor parallel clusters, vLLM / TensorRT-LLM runtimes, dynamic batching, and hardware-accelerated model routing.
About the role
AI models are only as good as the infrastructure serving them. You will ensure Cehpoint's proprietary models and agent clusters run with rock-solid reliability, sub-15ms time-to-first-token, and peak GPU utilization across cloud and on-premise hardware deployments.
Lead the infrastructure layer powering Cehpoint's AI compute. You will engineer ultra-efficient model serving clusters, optimize GPU memory bandwidth, configure dynamic batching, and maintain 99.99% uptime across distributed inference infrastructure.
What you will do
- Deploy, scale, and optimize inference engines (vLLM, TensorRT-LLM, TGI) across multi-GPU NVIDIA clusters.
- Implement PagedAttention, continuous batching, and speculative decoding to double token generation throughput.
- Configure automated model caching, weights distribution, and failover across regional edge locations.
- Design comprehensive Prometheus/Grafana monitoring dashboards tracking KV cache usage, queue depths, and token latency.
- Maintain rock-solid security hardening across cloud VPS instances, container runtimes, and VPC networks.
What you should know
- GPU microarchitecture, CUDA execution hierarchy, high-bandwidth memory (HBM), and NVLink interconnects.
- KV-cache mechanics, tensor parallelism, pipeline parallelism, and weight quantization techniques.
- Linux kernel networking, container isolation, and enterprise security compliance.
What you should be able to do
- Expertise in Linux system administration, bash automation, and infrastructure-as-code.
- Deep practical mastery of vLLM, Triton Inference Server, or TensorRT-LLM in production environments.
- Strong containerization experience (Docker, Kubernetes/K3s) and network routing (Nginx, Envoy).
How you work
- Passionate about squeezing maximum FLOPS and memory bandwidth out of modern hardware.
- Zero tolerance for server downtime, resource leaks, or unmonitored infrastructure degradation.
- Builder who loves testing newly released inference optimizations the day they drop.
Skills we check
What you get
- Generous annual package ₹12,00,000–₹21,00,000 with infrastructure uptime bonuses.
- Hands-on ownership of enterprise GPU hardware clusters.
- Remote-friendly team structure with flexible hours.
- Sponsorship for advanced Kubernetes and cloud architecture certifications.
How we select for this role
Every stage happens in the portal, and you can see where you stand at any time.
Apply on the portal
Sign in, upload your CV (PDF or Word), confirm your details and sign the candidate undertaking. About ten minutes.
CV review
Your CV is read against this role's requirements, usually within a few minutes. If it is not a fit, you see that in the portal with the reason.
Written questions and integrity check
Around ten technical questions written for your CV and this role, answered in the portal with paste turned off. An automated check confirms the answers were typed by you, and you see the result before you submit.
Shortlisted: choose your date
If no cheating is detected you are shortlisted straight away, and we email you a link to choose the day your practical project starts.
5-day practical project
On that day our team contacts you with the brief: real work for 5 working days, reviewed by an engineer. For this role: Set up an optimized multi-GPU inference engine achieving a 2.5x throughput improvement and sub-15ms time-to-first-token under synthetic concurrent load (strictly 5 working days).
Offer
If your work solves the problem well, your offer letter is issued in the portal and by email.
Other open roles in this team
- Agent Messaging & Orchestration EngineerSoftware & platform · Full-time · Remote (India)₹12,00,000 – ₹20,00,000 / year
- Agentic Skill Engineer (skill.md & Workflows)Software & platform · Full-time · Remote (India)₹10,00,000 – ₹18,00,000 / year
- GPU Infrastructure & MLOps EngineerSoftware & platform · Full-time · Remote (India)₹12,00,000 – ₹21,00,000 / year
- Lead Full-Stack AI Platform EngineerSoftware & platform · Full-time · Remote (India)₹9,00,000 – ₹16,00,000 / year
- Agentic Skill & Tool Synthesis InternSoftware & platform · Fellowship · Remote (India)₹15,000–₹22,000/month + Full-Time PPO (up to ₹14 LPA)
- Full-Stack AI Products Intern (Next.js / Node / Tailwind)Software & platform · Internship · Remote (India)₹10,000–₹18,000/month + Full PPO