MLOps.community Agentic Marketplace
37 snips
Mar 20, 2026 Donné Stevenson, ML engineer building scalable GenAI infrastructure and rollout strategies. Pedro Chaves, data science manager shaping GenAI search and marketplace recommendations. They explore agent-driven home search, enriched listings with commute and safety data, progressive permissioning and undo patterns, agent-to-agent marketplace layers, and agent-powered uploads, negotiation and logistics.
AI Snips
Chapters
Transcript
Episode notes
Home Search As A Lifestyle Experience
- Buyers search for homes as lifestyle events, not just properties, requiring agents to extract preferences about amenities, commute and neighbourhoods.
- Pedro built an interaction mixing chat and generated UI components to model personas and present enriched, commute-aware recommendations.
Roll Out Agents In Phased Betas
- Validate new agentic product flows via staged rollouts: internal teams, expert beta, invite-only public beta, then full launch.
- Collect explicit and implicit feedback and iterate quickly with elastic infra during each stage.
Shortcut Actions Bridge Dealers To Agents
- Dealers resist raw chat because they don't know what to ask or fear unwanted actions; prebuilt shortcuts and actions reduce friction.
- Donné found shortcut buttons and prefilled actions help dealers understand agent capability and follow with natural questions.