DevOps Engineer (Founding Team)

Role Overview

You will be the guardian of infrastructure and velocity. Every AI artifact, every mini-app, every TutorAgent interaction depends on the reliability of the pipelines and cloud environment you build. At Vibecoderz, DevOps isn’t a support role — it’s the backbone of product velocity and reliability.

As the founding DevOps engineer, you will design our CI/CD pipelines, observability stack, infra-as-code, and security posture from day zero. You’ll ensure that when a developer ships code, it’s live in production safely, quickly, and traceably. You’ll architect the serverless-first cloud strategy on GCP, balancing performance, cost, and scale as we grow from MVP to 100K+ MAUs.

You’ll use Linear (execution), Notion (runbooks/docs), and GitHub (actions & infra code) to make every workflow reproducible, transparent, and automated.

Key Responsibilities

1. CI/CD Pipeline Ownership

   • Design and maintain GitHub Actions workflows for FE, BE, AI, and agent services.

   • Automate build, test, and deploy to Cloud Run with zero-downtime releases.

2. Infrastructure as Code (IaC)

   • Implement Terraform scripts for GCP (Cloud Run, Firestore, Pub/Sub, VPCs).

   • Maintain environment parity (dev, staging, prod).

3. Observability & Monitoring

   • Set up OpenTelemetry tracing for multi-agent workflows.

   • Configure dashboards (Cloud Trace, Grafana) for latency, errors, and throughput.

4. Cost & Resource Optimization

   • Track infra costs, optimize workloads, and enforce scaling policies.

   • Benchmark agent workloads across Gemini Flash vs. Pro vs. custom models.

5. Cloud Security & Compliance

   • Enforce IAM best practices, firewall rules, and secret management.

   • Build guardrails for prompt injection and unsafe agent actions at the infra level.

6. Release Management

   • Define release pipelines with feature flags, rollbacks, and canary deploys.

   • Ensure smooth collaboration between PM, engineers, and QA.

7. Disaster Recovery & Resilience

   • Build automated backup and recovery strategies for Firestore + Neo4j + Redis.

   • Design failover strategies for critical agent services.

8. Agent Infrastructure Support

   • Support Browser-Use scaling for the Vibe Browser.

   • Manage GPU/TPU allocations for Gemini/Vertex pipelines if required.

9. Collaboration & Enablement

   • Write runbooks and incident playbooks in Notion.

   • Train engineering team to self-serve common workflows.

10. Problem Solving

    • Debug infra bottlenecks, trace latency across services, and enforce SLAs.

Success Metrics

90 Days (Probation):

• CI/CD pipeline live for FE + BE services.

• Terraform-based infra deployed and reproducible.

• OpenTelemetry traces visible for at least 2 core user flows.

12 Months:

• 99.9% uptime across production workloads.

• <200ms latency for API responses across multi-agent workflows.

• Fully automated deployments with rollback & feature flag system.

• Disaster recovery tested with <5 min RTO (Recovery Time Objective).

Must-Haves

• 10+ years in DevOps/SRE roles for high-scale products.

• Mastery of CI/CD, Terraform, GCP services (Cloud Run, Pub/Sub, Firestore).

• Proven experience with observability stacks (OpenTelemetry, Prometheus, Grafana).

• Deep knowledge of cloud security, IAM, and infra cost management.

• Background in scaling infra for developer or AI products.

Nice-to-Haves

• Experience with Vertex AI/ML infra and GPU/TPU scaling.

• Prior work on real-time, multi-agent systems.

• Contributions to open-source DevOps tooling.

• Startup/founding engineer experience.

Tech Stack Visibility

• Infra: Terraform, GCP (Cloud Run, Pub/Sub, Firestore, VPC)

• CI/CD: GitHub Actions

• Observability: OpenTelemetry, Cloud Trace, Grafana

• Security: IAM, Secrets Manager, GCP Firewall

• Other: Redis, Neo4j, Browser-Use scaling infra

Assessment

Objective: Validate ability to design and operate production-grade infra for Vibecoderz.

Challenge (Candidate PoC):

1. CI/CD Setup

   • Create a GitHub Actions workflow to:

     • Run unit tests for FE (Next.js) + BE (FastAPI).

     • Deploy BE service to Cloud Run on merge to main.

2. IaC

   • Write Terraform scripts to provision:

     • Cloud Run service

     • Firestore DB

     • Pub/Sub topic for agent comms

3. Observability

   • Add OpenTelemetry traces for one workflow (Text → Course).

   • Export traces to Cloud Trace and provide a screenshot of latency breakdown.

4. Security

   • Configure IAM policy with least-privilege roles.

   • Add secrets management (e.g., API keys) to the workflow.

Deliverables:

• GitHub repo with workflows + Terraform configs.

• Cloud Run URL for deployed BE service.

• Tracing screenshot with latency insights.

• Short README explaining infra choices + tradeoffs.

Evaluation Criteria:

• CI/CD Workflow Robustness (25%)

• IaC Quality & Reproducibility (25%)

• Observability & Monitoring Depth (20%)

• Security & IAM Best Practices (15%)

• Documentation & Clarity (15%)