Senior AI Integrations Engineer
Role Overview
Everyone's talking about AI agents. Almost no one is building the plumbing that makes them actually work. That's this job. Our Client, a heavyweight in OT/ICS cybersecurity, has business teams shipping no-code AI agents across the company. Those agents are only as good as what they can reach: clean warehouse data, secure system connections, APIs that don't fall over. You're the engineer who builds that layer. Model Context Protocol (MCP) servers. Production APIs into CRM and ERP systems. A Google Cloud architecture built on Vertex AI that scales. This is leverage, not maintenance. Every connector you ship gets reused across dozens of agents and hundreds of users. When a business team hits the wall of what no-code can do, you're the person they call. And you're the one who turns their prototype into production. Build and maintain MCP servers that expose internal systems, data, and tools to AI platforms: secure, governed, and logged Design and ship REST APIs that agents call to read from and write to CRM, ERP, and other systems of record Own the backend AI architecture on Google Cloud: Vertex AI, Agent Builder, Apigee, Cloud Run, Workflows Build the data pipelines that keep the warehouse stocked with clean, structured, AI-ready data Turn business users' agent prototypes into production-grade workflows, with authentication, authorization, observability, and data governance baked in
The Mission
- ·Build and maintain MCP servers that expose internal systems, data, and tools to AI platforms: secure, governed, and logged
- ·Design and ship REST APIs that agents call to read from and write to CRM, ERP, and other systems of record
- ·Own the backend AI architecture on Google Cloud: Vertex AI, Agent Builder, Apigee, Cloud Run, Workflows
- ·Build the data pipelines that keep the warehouse stocked with clean, structured, AI-ready data
- ·Turn business users' agent prototypes into production-grade workflows, with authentication, authorization, observability, and data governance baked in
Relevant Skills and Experience
- ·5+ years of software or data engineering, with strong Python and/or TypeScript
- ·Real data engineering chops: SQL, dimensional modeling, ETL/ELT, and hands-on work in a modern warehouse (BigQuery, Snowflake, Databricks, or Redshift)
- ·Deep Google Cloud fluency: Cloud Run or Cloud Functions, Apigee, Secret Manager, Cloud IAM
- ·Integration scar tissue with at least one major CRM and ERP (their APIs, data models, and quirks), plus experience shipping REST APIs other teams depend on
- ·Auth done right: OAuth 2.0, OIDC, service accounts, Workload Identity Federation
- ·Bonus points: you've built MCP servers before, or shipped RAG with vector search and embeddings (Vertex AI Vector Search, AlloyDB, BigQuery vector), or you know your way around PII handling and data governance
Job Summary
VMG guarantees strict confidentiality for all applicants.