Cache smarter, respond faster – Accelerating Conversational AI with Semantic Caching
Job Board за IT общността
Спонсори на събитието:
За събитието
Това събитие е част от потребителската група AI/ML/Big Data.
ПРОГРАМА:
Conversational AI systems often process every user query from scratch. But what if the system has answered this question before – only worded slightly differently – and all the hardest computation work has already been done? This talk explores leveraging semantic caching to overcome that inefficiency by focusing on:
>>> Dynamic Smart Reuse: Designing systems that recognize patterns and learn from interactions rather than just processing queries
>>> Speed Without Compromise: Navigating the trade-offs of fast-path retrieval, caching mechanics, and keeping the critical path clear on a miss.
>>> Adaptive Invalidation: Detecting semantic contradictions when new information arrives and dynamically invalidating outdated entries.
>>> Q&A
Събитието е безплатно, но със задължителна предварителна регистрация.
ПОТРЕБИТЕЛСКА ГРУПА: AI/ML/Big Data
Събитието е част от потребителската група за AI/ML/Big Data. Всеки месец организираме събития на тема свързана с AI/ML/Big Data. За да не изпускате интересна информация по темата – абонирайте се.
ОРГАНИЗАТОР: DEV.BG е специализиран Job Board за IT обяви.
AI/ML/Big Data обяви: Ако си търсиш работа свързана с AI/ML/Big Data, разгледай обявите тук.
КОГА: 05.10.2026 г. от 19:30 ч.
Очакваме ви!
За лектора
Rosen is a Senior Machine Learning Engineer with over 10 years of experience designing and building production-grade SaaS products using Python and C++ back in the days. His work spans applied machine learning, large-scale system architecture, and end-to-end AI deployment in real-world enterprise environments, with a strong focus on reliability, scalability, and maintainability. Throughout his career, he has worked on delivering ML-powered features from research and prototyping all the way to robust, customer-facing production systems.
In recent years, Rosen has specialized in deep translation, multilingual NLP, intelligent document processing, data extraction pipelines, and OCR-driven document understanding. He is particularly interested in solving challenges around low-resource languages, technical terminology, and domain-specific translation quality. He combines strong engineering foundations with practical ML expertise, bridging advanced model capabilities with real product constraints to deliver systems that perform consistently in demanding enterprise settings.
За лектора
Martina is a Machine Learning Engineer at Iris.ai, where she works on building AI-powered applications, with a focus on RAG and agentic AI systems. She has around six years of experience in technology, spanning full-stack web development, AI consulting, and machine learning engineering. Her work focuses on taking AI solutions from experimentation and prototyping towards reliable, practical systems that can be used in real-world applications.
Нашите
Партньори
Iris.ai е вашият най-добър помощник в научните изследвания, който използва NLP, за да намира, извлича и обобщава ключови данни от огромни научни бази данни.
Бъди част от AI/ML/Big Data общността
IT Събитие на фокус
