LLM providers are trying to build enterprise guardrails directly into their products, but it won't work. Companies don't run on a single LLM any more, they're often using a mix of frontier models plus open-source and specialized vertical models spread across various use cases. Model providers can improve their guardrails, but that won't fix the fragmentation problem. The industry tried this before! In the early microservices era, every team baked auth, rate limiting, and logging directly into their apps, which worked fine until you had hundreds of services and no consistent enforcement or unified observability. We abstracted the connectivity logic to the traffic layer back then, and you need to do it again with AI. Effective AI governance (token budgets, permissions, compliance, etc) can't live inside a single model. You need a neutral layer that sits in front of everything. A "Switzerland for AI" as Khozema Shipchandler called it. A single control tower that manages all traffic regardless of source.

LLM Knowledge Bases Something I'm finding very useful recently: using LLMs to build personal knowledge bases for various topics of research interest. In this way, a large fraction of my recent token throughput is going less into manipulating code, and more into manipulating knowledge (stored as markdown and images). The latest LLMs are quite good at it. So: Data ingest: I index source documents (articles, papers, repos, datasets, images, etc.) into a raw/ directory, then I use an LLM to incrementally "compile" a wiki, which is just a collection of .md files in a directory structure. The wiki includes summaries of all the data in raw/, backlinks, and then it categorizes data into concepts, writes articles for them, and links them all. To convert web articles into .md files I like to use the Obsidian Web Clipper extension, and then I also use a hotkey to download all the related images to local so that my LLM can easily reference them. IDE: I use Obsidian as the IDE "frontend" where I can view the raw data, the the compiled wiki, and the derived visualizations. Important to note that the LLM writes and maintains all of the data of the wiki, I rarely touch it directly. I've played with a few Obsidian plugins to render and view data in other ways (e.g. Marp for slides). Q&A: Where things get interesting is that once your wiki is big enough (e.g. mine on some recent research is ~100 articles and ~400K words), you can ask your LLM agent all kinds of complex questions against the wiki, and it will go off, research the answers, etc. I thought I had to reach for fancy RAG, but the LLM has been pretty good about auto-maintaining index files and brief summaries of all the documents and it reads all the important related data fairly easily at this ~small scale. Output: Instead of getting answers in text/terminal, I like to have it render markdown files for me, or slide shows (Marp format), or matplotlib images, all of which I then view again in Obsidian. You can imagine many other visual output formats depending on the query. Often, I end up "filing" the outputs back into the wiki to enhance it for further queries. So my own explorations and queries always "add up" in the knowledge base. Linting: I've run some LLM "health checks" over the wiki to e.g. find inconsistent data, impute missing data (with web searchers), find interesting connections for new article candidates, etc., to incrementally clean up the wiki and enhance its overall data integrity. The LLMs are quite good at suggesting further questions to ask and look into. Extra tools: I find myself developing additional tools to process the data, e.g. I vibe coded a small and naive search engine over the wiki, which I both use directly (in a web ui), but more often I want to hand it off to an LLM via CLI as a tool for larger queries. Further explorations: As the repo grows, the natural desire is to also think about synthetic data generation + finetuning to have your LLM "know" the data in its weights instead of just context windows. TLDR: raw data from a given number of sources is collected, then compiled by an LLM into a .md wiki, then operated on by various CLIs by the LLM to do Q&A and to incrementally enhance the wiki, and all of it viewable in Obsidian. You rarely ever write or edit the wiki manually, it's the domain of the LLM. I think there is room here for an incredible new product instead of a hacky collection of scripts.

// Your LLM judge disagrees with the experts // LLM Judges can be tricky to build. Here is an interesting showcasing why: There propose a reference-full benchmark of hundreds of complete human-to-human dialogues written by professional script writers, with realistic turn densities and more than 36,000 per-turn human annotations across over 30,000 expert-generated turns. Conversational evaluation frameworks were mostly built for summarization, translation and short-form QA, and the metrics themselves are often derived and validated on synthetic data rather than human dialogue. Tested against expert judgment at this scale, both classical automatic metrics and reference-free LLM-as-a-judge approaches turn out to be unreliable. Their Mixture-of-Judges framework combines multiple evaluative signals and recovers roughly 30 percent better correlation with human assessment. Paper: Chat with Paper:

every LLM provider has the option of advertising cheap sticker prices but their cache ratios can be bad so your effective price is higher the root thing thing though is idk if 100s of providers advertising to end users makes sense long term