Semantic Caching in RAG Applications
Imagine you’re running an application that uses a Large Language Model (LLM) like GPT-4, Anthropic Claude, etc. to answer customer questions. Every time a user asks something, your system sends a q...
Keyur Ramoliya
Tesla's Open-Source Patent Strategy
In 2014, Tesla made a big decision to share its patents with everyone. This move, which allowed other companies to use Tesla’s technology without paying for it, surprised many in the industry. Let’...
SpreadsheetLLM: Encoding Spreadsheets for Large Language Models
Spreadsheets have long been a fundamental tool in data management and analysis. However, their complex structures have posed significant challenges for AI systems, particularly large language model...
BMW Connected Vehicle and App Experience Architecture
Cloud-native architectures are always interesting. Recently, in Microsoft Build 2024, BMW shared how their mobile apps are powered by Microsoft Azure. Engineers from BMW have shown how their app ...
Test-Time Training (TTT): A New Approach to Sequence Modeling
In artificial intelligence (AI) and natural language processing (NLP), sequence modeling is incredibly important. It’s the foundation for many applications we use every day, like language translati...