All interviews
Figma logo

Figma

Mid

Build the integrations and AI workflows behind Figma's AI-powered support experience

Figma's AI Infrastructure & Tooling team needs an engineer to connect Decagon, Zendesk, Figma admin tooling, and internal data sources so AI chatbots and support Specialists get the right context automatically. The work spans back-end integration engineering (APIs, webhooks) and applied LLM patterns (classification, routing, summarization) with production guardrails. Expect the interview to test both hands-on integration-building skill and judgment about safe, measurable AI rollout in a customer-facing operational system.

Step into this interview

Free · a live voice mock calibrated to this exact role

Practice this interview

What this interview tests

  • Cross-system integration engineering (Decagon, Zendesk, internal admin tooling)
  • Back-end proficiency: APIs, webhooks, data pipelines (Ruby/Python/Go/PostgreSQL)
  • Applied LLM patterns for support: classification, routing, summarization, context enrichment
  • Production guardrails: monitoring, fallback paths, quality checks for AI workflows
  • Defining and tracking success metrics for support automation (containment, deflection, CSAT, FCR)
  • Translating ambiguous support problems into scoped technical solutions

Common question themes

Walk me through an integration you built end-to-end between a support platform and an internal system

How would you design context enrichment so a support chatbot has the right account/billing metadata?

Describe an LLM-powered workflow you shipped for classification, routing, or summarization and how you measured its impact

How do you build fallback paths and guardrails so an AI support workflow is safe in production?

How would you define success metrics for a new AI-powered support automation?

How do you scope an ambiguous support problem into a concrete technical solution?

View the original posting

Related interviews