21 Aug
|
HCLTech
|
Bengaluru
AI PRE Engineer (Platform Reliability / Production Readiness Engineer)
The Role
An AI PRE Engineer ensures AI/ML platforms are production-ready, highly reliable, observable, secure, and cost-effective, bridging AI engineering, SRE, DevOps, and MLOps disciplines.
Responsibilities:
- Define and maintain production readiness standards across platform, data, model, application, and security layers.
- Manage the NVIDIA Enterprise AI Software stack and deployment of GPU enabled Kubernetes clusters.
- Establish SLO/SLI frameworks for latency, availability, quality, safety, and drift implement error budget policies.
- Implement and configure the LLM inferencing services using kServe, NIM, AI Gateways
- Curate deployment blueprints (canary/shadow, blue–green, A/B) for models and prompts with rollback guidance.
- Standardize observability patterns for prompts, embeddings, latency, cost, quality, and safety telemetry.
- Own capacity engineering (token/concurrency budgets, GPU/CPU sizing, vector scaling, cache hierarchies).
- Define resilience patterns (timeouts, circuit breakers,
fallbacks, idempotent retries, semantic/prompt caching).
- Set AI security baselines (secrets, private networking, egress controls) and mandate red‑team & safety evaluations.
- Maintain compliance mappings (e.g., ISO 27001, SOC 2, GDPR/DPDP, HIPAA where applicable).
- Provide CI/CD pipelines, SDKs, Helm/Terraform templates, and policy‑as‑code for consistent delivery.
- Author PRR checklists, runbooks/playbooks, and DR/BCP blueprints (RTO/RPO, multi‑region/site failover). Drive enablement (trainings, brown-bags) and maintain knowledge repositories and decision records.
- Partner with solution teams to validate architecture and non‑functional requirements (scale, latency, cost, safety).
- Conduct Production Readiness Reviews (PRRs) and certify releases across performance, security, privacy, and compliance.
- Implement observability (tracing, metrics, logs), dashboards, and SLO burn and cost anomaly a
📌 Platform Engineer (Bengaluru)
🏢 HCLTech
📍 Bengaluru