Senior Technical Program Manager (Services Tools & Diagnostics)

Role overview

Join Twilio as a Senior Technical Program Manager (Services Tools & Diagnostics) in the GCS Strategy & Operations organization. This is a hands-on role designing and building systems that power GCS service delivery workflows. You’ll translate business requirements into reliable, production systems, with success measured by customer outcomes.

Key missions

• Architect and build systems that encode how GCS operates, including:
  • Account health observability
  • Intelligent routing
  • Knowledge retrieval
  • Workflow automation
• Design and build AI workflows using:
  • Anthropic Claude API
  • Claude Agent SDK
  • Claude Code
  • MCP servers to connect agents to internal systems
• Partner with CXO stakeholders to create decision frameworks for when systems act autonomously, ensuring guardrails to keep failure rates low

Responsibilities / expectations

• Turn business requirements into clear system specifications
• Define metrics, observability, and evaluations for AI systems in production
• Run structured A/B tests or evaluations for reliability and performance
• Build production AI capabilities (tool use/function calling, structured outputs, prompt architecture, evaluation harnesses)
• Develop production-grade integrations and APIs for enterprise systems

Requirements

• 7+ years building production software/technical systems (e.g., platform engineering, solutions architecture, or hands-on technical program management)
• Advanced SQL, including analytical queries against columnar stores (e.g., ClickHouse)
• Deep understanding of large language models (reasoning, failure modes) and RAG architectures
• Strong prompt engineering patterns and agent frameworks
• Experience embedded in support or professional services, building systems for teams that deliver the work
• Ability to explain complex technical behavior to non-technical stakeholders
• Experience defining AI observability metrics and running A/B tests or evaluations
• Hands-on building and deploying AI systems with LLMs, including MCP servers and enterprise integrations
• Familiarity with Twilio products and/or the CPaaS/SaaS domain
• Experience with Zendesk (APIs, triggers, routing logic)
• Strong engineering practices:
  • Git-based version control and code review
  • Basic CI
  • Least-privilege patterns for customer data
• Production-quality proficiency in Python and/or TypeScript (Node.js)
  • REST fundamentals: authentication (OAuth 2.0, API keys, token refresh), webhooks, pagination, rate limiting, error handling, retries, and JSON transformations
• Dashboard/application experience across one or more of:
  • Grafana
  • Salesforce (SOQL, report types, Lightning reports/dashboards)
  • Tableau (data modeling, calculated fields, LOD expressions; publishing to Tableau Cloud/Server)
  • Looker (LookML modeling, Explores, derived tables)
  • Airtable (base/table design, automations, scripting extensions, API integration)

Nice-to-haves

• Breadth across multiple BI/service platforms (Grafana/Salesforce/Tableau/Looker/Airtable) and ability to choose the right tool based on audience and data source