Interview with TomTom's
summary
Cloud-hosted in-vehicle assistant uses Azure Cosmos DB with the Python SDK to store conversation history and evolving context, enabling rapid prototyping and on-the-fly schema changes typical of NoSQL. The team integrates retrieval-augmented generation for context retrieval, confronts non-deterministic answers that break traditional tests, and reports the assistant working in cars as a prototype while highlighting iteration speed and changed testing and engineering patterns as the main outcomes.
AI-written summary · about
related talks
47mMaking the CASE: Cloud-based innovation in the automotive industry
32mData&AI in TomTom: How are are transforming mobility
42mFrom AI Conversations to Data-Driven Decisions: Unlocking the Future of Customer Engagement
56mMake A Company Conversational using Azure Cognitive Services
46mMultimodel Multimodal and Multiagent innovation with Azure AI
01h 09mBuild your own KITT with Azure Cognitive Services
01h 03mTime Traveling in the Cloud
49mMake your applications intelligent (...)