Anapoly AI Labs

Tag: obsidian

  • Testing a local AI

    I am using Obsidian to build not a second brain, but a workspace for my brain: a space in which to think. The workspace is intended to become an ideas factory, designed as a knowledge network: a network of ideas at different stages of development.

    There is scope for AI technology to enhance my thinking in that space. For example, ideas in the knowledge network can be embedded into a vector database, roughly similar to how an AI organises knowledge. This allows similarities among widely scattered notes — the semantic connections — to be highlighted, and lets me search not just for specific words but also for ideas with related meaning. The Smart Connections plugin implements these useful capabilities.

    I can take this a step further by inviting an AI into my thinking space. Ollama lets me run a large language model on my laptop and connect it to my Obsidian vault through the Smart Connections plugin, so the AI can query, summarise, and respond directly from my notes. I downloaded the Mistral 7B Q4_K_M model for this purpose. Put simply, it’s a compressed version of the 7B model that runs faster and uses less memory, at the cost of some accuracy.

    I ran two tests. First, I pointed it to a short daily note and asked, “What is Claire’s new job?” After a pause of about half a minute, it gave me the correct answer.

    For the second test, I pointed it to a more substantial document and asked, “What is the purpose of Goal-Directed Context Management?” The reply took eight minutes, but the answer was very good:

    The purpose of Goal-Directed Context Management, as outlined in the provided document, is to manage a large language model (such as ChatGPT) effectively when it’s being used as a collaborator during the production or editing of knowledge-based products.

    This framework aims to configure and adapt the behavior of the AI to suit different roles over the course of work, while also providing scaffolding in the form of project and program management techniques to keep the AI focused and effective over the lifecycle of an information-based product.

    By aligning the AI’s evolving context with the purpose and structure of a project or program, Goal-Directed Context Management ensures that the AI always has the right information to contribute effectively to the current stage of work.

    The test confirmed three things: I could run an AI locally to keep my data private, the model gave accurate responses, and, although performance was slow on my three-year-old Dell XPS, it still worked. Watching my own laptop pull a thoughtful answer from my notes was oddly satisfying, even if I had time to make tea while it thought.

  • Mind Maps, Podcasts, and a Pocket Brain

    Lately I’ve been testing a simple idea: can AI turn Obsidian into a genuinely useful second brain for me? The answer was a complicated but resounding “Yes”.

    My first step was to find an experienced Obsidian user whose ideas resonated with mine. This turned out to be a YouTuber called Callum (aka Wanderloots). Callum’s channel has over fifteen  videos relating his personal experience in building a second brain, and offering advice about all aspects of Obsidian for that purpose. 

    I satisfied myself that Callum’s approach would be a good basis from which to develop my own, but didn’t have the time to watch every video in order to benefit from their content. I needed a quick and efficient way to fast-track that process. Step forward NotebookLM.

    One of the great things about NotebookLM is that you can give it fifteen YouTube videos and then have a conversation about their content. The discussion can encompass the content of one, several, or all of the videos. To help you structure the conversation, NotebookLM can produce a mind map setting out all the concepts or ideas contained in the videos. 

    On top of that, to help you reflect on these ideas while strolling round the park after work, the AI can produce an audio overview. This takes the form of a podcast-style discussion between two hosts, and you can set the ground rules for their discussion, for example the focus points, audience, technical level. Listen in for yourself.

    Intriguingly, the discussion is interactive when you’re online to the AI. You can join in to ask questions or steer the discussion in a particular direction.

    With the big picture in place, the next step was the hands-on work of shaping Obsidian to fit my needs. That will be the subject of my next post, where I’ll dig into the practicalities of building it and explore how a local AI might give my second brain extra intelligence without compromising its privacy.