Posts tagged opinions

Implementing Semantic Caching in Production

Caching is a core strategy for optimizing performance of computation-intensive applications. While this approach reduces computational overhead thus cost and latency in theory, productionizing semantic caching requires careful consideration of possible limitations.

Read more ...


My Two Cents on Agentic AI

Agentic AI has become a popular buzzword in our field. While it has been claimed to be very promising, what it entails in terms of practical implementations and best practices is still very new and developing. As I dig deeper, I’ve found that the shapes and forms of agentic workflows are not as clear as I thought, even though the internet is flooded with articles and videos on the topic. This post offers a cleaner definitional abstraction of the idea, some examples, and some honest opinions on the topic.

Read more ...