Category: Blog

  • PGVector Retriever DSpy

    Below you can find a simple Notebook to use PGVector with a custom retriever in DSpy. See the Github link too: https://github.com/tulip-ai/notebooks

  • The need for programmatic prompt engineering

    In the rapidly evolving field of AI and natural language processing, tools like Langchain and Haystack have made significant strides in building pipelines for large language models (LLMs). However, these solutions often fall short in one crucial area: providing a truly programmatic way to create retrieval-augmented generation (RAG) pipelines. To unlock the full potential of…

  • RAG

  • Are chatbots essential?

    No doubt you’ve heard of LangChain and perhaps even played with it, seeing how easy it is to create your own LLM-powered chatbot. If you haven’t, here’s a brief intro. LangChain was launched in 2022 as a groundbreaking tool leveraging the power of large language models to create chatbots. One of the most remarkable features…

  • To train or not to train

    Should you use your company data to train a LLM? And is there any benefit in this vs just fine-tuning an existing one? There’s a prevailing lack of understanding around what exactly “training” is and overexcitement to get models deployed and running in businesses can lead to all sorts of problems. Executive Excitement I was… …