How modern search is built and where LLMs fit in.How semantic search architectures are designed and evaluatedWhy BERT is still the baseline in many systemsFine-tuning strategies, training data, and infrastructure trade-offsWhat's possible in multimodal, and what usually isn't worth the effortChoosing between PostgreSQL and vector databasesWhat often blocks companies from improving search, from training data to infra🥐 Hosts - Anton Zhuravsky & Dmitry Volokh☕️ Guest – Roman Grebennikov , who has been building search at Delivery Hero, and has also built his own search engine – Nixieserach.aiThe podcast is brought to you by ⁠On The Spot Development⁠, a software development company which Anton founded. They build R&D teams for startups and tech companies.

Ver todos los episodios