Senior AI Platform PM – Observability & Automation

SendGrid; United States; On-site; USD 199,280 - 249,100; Full time

Job summary

Twilio is seeking a highly skilled engineer to architect and build scalable AI-enabled systems for GCS operations. You will implement observability, intelligent routing, and workflow automation while optimizing costs across TAM, PS, and EM motions.

You will create AI workflows using Claude APIs and MCP servers that connect to enterprise tools like Slack, Jira, Zendesk, Grafana, and Airtable, ensuring guardrails and low failure rates.

Qualifications

• 7+ years building production software or technical systems in platform engineering, solutions architecture, or technical program management.
• Proficiency in Python and/or TypeScript with REST API fundamentals: auth, webhooks, pagination, error handling, retries, and JSON transformation.
• Hands-on experience building AI systems with LLMs — tool use, function calling, prompt architecture, evaluation harnesses; MCP servers for enterprise connections.
• Deep understanding of LLMs: reasoning, failure modes, RAG, prompt engineering, agent frameworks.
• Advanced SQL including analytic queries on ClickHouse.
• Dashboard/app development across Grafana, Salesforce, Tableau, Looker, Airtable; selecting appropriate tools for audiences.
• Strong engineering practices: Git, code reviews, CI, least-privilege data access.
• Ability to translate business needs into system specs and communicate with non-technical stakeholders.

Responsibilities

• Architect and build systems for GCS operations: observability, routing, knowledge retrieval, and workflow automation; optimize costs.
• Develop AI workflows using Claude ecosystem and MCP servers linking to Slack/Jira/Zendesk/Grafana/Airtable and internal Twilio systems.
• Collaborate with CXO to design autonomy decision frameworks and guardrails to minimize failure rates.
• Enhance observability with dashboards across Grafana, Salesforce, Tableau, Looker, Airtable; choose appropriate tools for audiences.
• Assist with governance maturation: versioning, evaluation frameworks, AB testing, performance measurement for GCS.
• Instrument systems to measure retrieval accuracy, action success, deflection, failure patterns; tie to outcomes like time-to-value and cost-per-ticket.
• Translate complex system behavior into plain language for directors; produce clear technical docs for teams.

Skills

Python

TypeScript (Node.js)

REST API integration

AI systems with LLMs

MCP servers

RAG architectures

SQL (ClickHouse)

Dashboard tools

Tools

Claude API

Claude SDK

Claude Code

Job description

Twilio is seeking a highly skilled engineer to architect and build scalable AI-enabled systems for GCS operations. You will implement observability, intelligent routing, and workflow automation while optimizing costs across TAM, PS, and EM motions.

You will create AI workflows using Claude APIs and MCP servers that connect to enterprise tools like Slack, Jira, Zendesk, Grafana, and Airtable, ensuring guardrails and low failure rates.