---
session_id: "1352199"
title: "Building a Shared AI Assistant: The Platform Challenges Beyond MCP"
phase: live
last_checked: 2026-11-12T13:24:21-08:00
render: 20261112-132421-c4d6ee
---

# Links: Building a Shared AI Assistant: The Platform Challenges Beyond MCP

## Speaker

**Utkarsh Shrivastava**, engineering lead for enterprise AI, Netflix

- LinkedIn: [Utkarsh Shrivastava - Netflix](https://www.linkedin.com/in/utkarshshrivastava)

From the conference roster; the page's title names him and Netflix.

## Company

**[Netflix](https://www.netflix.com/)**: streaming service for movies and TV
shows. The talk's example is Netflix's own internal assistant ("Netflix
launched one of our").

## Resources

- **Blog** · [Parsing “Can” from “Should”: How Netflix's Product Team Got Thousands of Creatives to Trust an AI Overhaul](https://review.firstround.com/netflix-ai-strategy/)
  (First Round Review, 2026-09-23). Netflix's internal AI rollout from the
  organisational side: the same problem the talk treats as a platform
  question.
- **Video** · [Scaling AI agent development at Netflix: Production insights with Claude Sonnet 4.5](https://claude.com/resources/webinars/scaling-ai-agent-development-at-netflix)
  (Anthropic webinar). Netflix engineering leaders on running AI agents in
  production. Related background, not this talk.

## Gaps

- The Netflix Tech Blog (netflixtechblog.com) answered 403 to the link check,
  and no post on the internal assistant or its MCP gating was found.
- An older profile (The Org) lists a data-infrastructure role; the roster's
  enterprise-AI role is used here.
- No official recording or slides found yet. The dossier rechecks every 6 hours
  during the event.
