Hi everyone,
Sharing a resource that might be relevant to the group's work.
Put together a detailed walkthrough of how end-to-end Generative AI applications are actually designed and built from an industry standpoint — covering the complete architecture from interface to deployment, including retrieval, orchestration, guardrails, evaluation, and observability.
The focus is on how these components connect in practice, not just in theory — which might be useful context for anyone working on applied GenAI research or thinking about how academic work translates into production systems.
Video link: https://www.youtube.com/watch?v=YsrNtMuYWXU
Free architecture slides: https://docs.google.com/presentation/d/e/2PACX-1vTMJva38eTJzptdbqBhtDETYVS0qnBSqzxLCZ62lWL-yw7uKhQnPp06ZGMkTDfR6qNGHUcNfwbHJ6Xz/pub?start=false&loop=false&delayms=3000&slide=id.p6
Happy to discuss any of the concepts or answer questions from the group.
Best regards,
Anuj Saini