Infrastructure Platform Engineer

• Has shipped infrastructure for at least one AI or LLM product to production.
• Owns the infrastructure for a RAG pipeline.
• Has managed competing infrastructure priorities across two or more product teams simultaneously.
• Proven ability to self-organize across parallel programs without a dedicated PM per track.
• Has been the primary technical owner in an InfoSec or security review for a customer-facing product.
• Can prepare evidence, own technical security questions, and drive review to sign-off without coaching.
• Has on-call experience for a production service with an SLA.
• Has been paged, resolved incidents, and written post-mortems.

• Experience ( Relevant) : 5-8 years in a platform engineering, SRE, or infrastructure engineering role

• Kubernetes / Containers / Helm
• Azure Platform (DevOps + Monitor)
• Terraform (IaC)
• GitOps (ArgoCD / Helm)
• CI/CD (GitHub + Azure DevOps)
• Observability (Otel / ELK / Datadog)
• SRE (on-call, RCA, SLO/SLA)
• Security / InfoSec readiness
• Vector DB / embeddings infra
• RAG pipeline (production)
• GPU infrastructure (A100/H100)
• LLM inference (Triton / NIM)
• End-to-end AI infra ownership
• Infrastructure Platform Engineer
• Let’s look for 2 resources one onsite and another offshore (the offshore can be a Infra Platform engineer with backend engineer skills)
• Manages everything below the application layer — compute, orchestration, RAG pipeline, CI/CD, observability, load testing, and security posture
• 5-8 years in a platform engineering, SRE, or infrastructure engineering role — infra-primary, not full-stack with infra responsibilities

Container Orchestration & Runtime

• Kubernetes / EKS – Production cluster operations, workloads, RBAC – Expert (Must-have)
• Docker – Image builds, multi-stage, container lifecycle – Expert (Must-have)
• Helm – Chart authoring, templating, release management – Advanced (Must-have)

Cloud Platforms

• Azure – Required
• AWS – Good to have
• Azure Key Vault – Secrets, certificates, managed identity binding – Advanced (Must-have)
• Azure Monitor + Log Analytics Workspace – Proficient (Must-have)
• Azure DevOps Pipelines – YAML pipelines, service connections, approvals – Proficient (Must-have)
• Azure Kubernetes Service (AKS) – Proficient (Nice to have)
• Azure API Management – Gateway, OAuth 2.0, rate limiting – Familiar (Nice to have)

Infrastructure as Code

• Terraform – Modules, state management, remote backends – Expert (Must-have)
• GitOps – ArgoCD or Flux CD, App-of-Apps pattern – Advanced (Must-have)
• Pulumi (Python / TypeScript) – Proficient (Nice to have)

AI / ML Infrastructure

• Vector database operations – Provisioning, indexing, tuning (Milvus, Weaviate, pgvector, FAISS) – Advanced (Must-have)
• GPU compute provisioning – NVIDIA data center GPU environments (B200, A100, H100) – Advanced (Must-have)
• RAG pipeline infrastructure – Ingestion, chunking, embedding pipeline – Proficient (Must-have)
• LLM inference infrastructure – Serving, scaling, latency management (NIM, Triton) – Proficient (Must-have)
• NIM deployment and lifecycle management – Proficient (Nice to have)

CI/CD & Deployment

• GitHub Actions – Workflow design, reusable actions, secrets – Advanced (Must-have)
• Azure DevOps Pipelines – YAML pipelines, approvals – Proficient (Must-have)
• Container registry management – Tagging, scanning, promotion (ECR, ACR) – Proficient (Must-have)

Observability & Monitoring

• Datadog – Metrics, APM, log management, dashboards, SLOs – Advanced (Must-have)
• Elasticsearch / OpenSearch – Indexing, querying, ILM policies – Proficient (Must-have)
• OpenTelemetry – Instrumentation, collector, exporters – Proficient (Must-have)
• Lightstep / Distributed tracing – Proficient (Must-have)
• Azure Monitor + Log Analytics integration – Proficient (Must-have)
• Incident response / On-call – PagerDuty or equivalent – Proficient (Must-have)

Performance, Security & FinOps

• Load testing – Scenario design, results analysis, SLA setting (k6, Locust, JMeter) – Advanced (Must-have)
• Security posture – Secrets hygiene, network isolation, RBAC audit – Advanced (Must-have)
• InfoSec review readiness – Controls documentation, evidence packaging – Proficient (Must-have)
• Python scripting for infrastructure automation and ingestion pipelines – Proficient (Nice to have)