Member of Technical Staff - Platform Engineering

Overview

Senior Platform Engineer to lead the "Inner Loop" of AI development, bridging high-velocity AI experimentation and enterprise-grade production infrastructure for LLM applications, RAG pipelines, and agentic workflows, as part of a dedicated AI Applications team.

What you'll do

• Design internal APIs and abstractions allowing engineers to provision AI-ready environments complete with model weights, vector databases, and event streams via single-command automation.
• Develop robust, type-safe internal tooling and AI agents in Go and Python to automate root-cause analysis, disaster recovery, and proactive infrastructure optimization.
• Architect asynchronous processing systems using Kafka and RabbitMQ to trigger dynamic infrastructure scaling and handle high-volume data streams like long-running document ingestion.
• Maintain "Golden Path" deployment standards using Docker and Kubernetes to keep models, prompts, and application code perfectly synchronized across all environments.
• Manage GPU workloads, distributed caches (Redis), and vector databases while monitoring system health and AI-specific metrics like token latency and model drift.

What you'll need

• 8+ years of experience in Platform Infrastructure Engineering with an emphasis on supporting AI application development and platform self-service.
• High proficiency in Golang.
• Hands-on experience integrating LLM APIs, prompt engineering, and RAG pipelines.
• Expertise in managing vector databases and distributed caches.
• Deep experience with Kubernetes.
• Deep experience with asynchronous message streaming via Kafka or RabbitMQ.
• Strong background in full-stack monitoring with Prometheus, Grafana, ELK to maintain operational resilience, high throughput, and zero-downtime standard operating procedures.

Details

• Location: Bangalore, India.
• Primarily an in-office environment; expected to work from the Bangalore office in compliance with Pure's policies, unless on PTO, work travel, or other approved leave.