AgenticOps Platform Engineer Lead

We are looking for a senior, hands-on AgentOps Platform Engineer to design, build, and operate the cloud-native infrastructure that powers our AI agents at scale.

This is a lead-by-example role:

• You write the Terraform
• You build the pipelines
• You own the platform in production

GCP is your primary environment, but you will design with multi-cloud in mind (AWS, Azure), ensuring portability, resilience, and long-term flexibility. This role sits at the intersection of DevOps, MLOps, and AgentOps, with deep responsibility for reliability, security, observability, and cost.

KEY RESPONSIBILITIES

Platform & Infrastructure Ownership

• Design, build, and operate production-grade infrastructure for AI agents and LLM services
• Own Terraform-based Infrastructure as Code for all environments (dev, uat, prod)
• Lead infrastructure decisions through hands-on implementation, not diagrams
• Build scalable foundations for: Agent orchestration Inference services RAG pipelines Vector stores
• Optimise cloud resources for performance and cost efficiency

AgentOps & AI Platform Enablement

• Enable safe, continuous operation of autonomous agents
• Design agent runtime environments with: Isolation & sandboxing Failover and recovery strategies Controlled rollout mechanisms
• Support prompt/version management, agent configuration, and tool/plugin lifecycle
• Work closely with Agentic RAG engineers to operationalise research into production

CI/CD & Automation

• Build and maintain CI/CD pipelines for: Infrastructure Agent services Prompt and config changes Model/version rollouts
• Automate workflows for: Vector DB updates RAG index refreshes Agent memory stores Tool registration and validation
• Reduce manual ops toil aggressively through automation

Observability & Production Readiness

• Design and implement deep observability for agent systems: Platform health Agent execution metrics Latency, cost, and throughput Failure modes and retries
• Build dashboards, alerts, and telemetry using: Prometheus Grafana OpenTelemetry (or equivalent)
• Enable visibility into agent decision traces and runtime behavior

Security, Safety & Reliability

• Implement secure cloud architecture and IAM best practices
• Own production reliability, incident response, and recovery
• Enforce operational guardrails and safety controls for agent APIs
• Support responsible AI practices from an infrastructure and runtime perspective

Collaboration & Technical Leadership

• Work closely with: Agentic RAG engineers AI engineers Product & CTO Office
• Define SLOs, reliability targets, and operational metrics
• Set the technical bar for AgentOps at BridgeAI
• Mentor engineers by example and code, not process overhead

REQUIRED SKILLS & EXPERIENCE

Core Platform & DevOps

• 5+ years in DevOps, Platform Engineering, SRE, or MLOps
• Strong, hands-on experience with GCP: GKE / Compute Engine Cloud Run / Functions Cloud Storage, Pub/Sub Vertex AI (or equivalent)
• Deep experience with Terraform (mandatory)

Containers, CI/CD & Automation

• Docker, Kubernetes, Helm
• CI/CD tooling (GitHub Actions, Jenkins, ArgoCD)
• Python and Bash for automation and platform glue code

Agentic & AI Systems

• Experience supporting LLM-based systems in production
• Understanding of: Prompt/version management Context handling & caching Model rollout strategies
• Hands-on experience with vector databases (Weaviate, FAISS, Pinecone)
• Familiarity with RAG pipelines and agent execution patterns

Observability & Security

• Monitoring and telemetry using Prometheus, Grafana, OpenTelemetry
• Strong understanding of cloud security, IAM, and operational safety

NICE TO HAVE

• Multi-cloud experience (AWS, Azure)
• Exposure to agent frameworks (LangChain, LangGraph, AutoGen, CrewAI)
• Event-driven systems (Temporal, Airflow)
• Experience with responsible AI operations or safety monitoring

WHAT SUCCESS LOOKS LIKE

• Infrastructure is reproducible, observable, and boring (in a good way)
• Agent failures are visible, debuggable, and recoverable
• Cloud costs are understood and controlled
• Engineers trust the platform and move faster because of it
• You are the go-to authority for AgentOps at BridgeAI

WHAT THIS ROLE IS (AND IS NOT)

• Deeply hands-on
• Terraform-first
• Production ownership
• Sets standards by building
• Not a people-manager role
• Not a ticket-based ops role
• Not a “just keep the lights on” job