Zach Solomon

Communications @awscloud. I enjoy exploring, fitness, nutrition, baseball, and playing guitar.

[Zach Solomon @ZachSB]

@MrBeast 4637283

[Zach Solomon @ZachSB]

@max_davish @evisdrenova Agreed. Same reason why forums / bulletin boards worked so well

[Zach Solomon @ZachSB]

@DrewVento I miss forums

[Zach Solomon @ZachSB]

I can’t remember the last meeting I had when I wasn’t running Kiro or Claude Code in the background. It’s absolutely becoming clear how AI is doing as I’m steering and aligning.

[Zach Solomon @ZachSB]

@aka60643 @cornoisseur whatever it is, would love some action to make it better.

[Zach Solomon @ZachSB]

@chiguypotts @cornoisseur To me that’s education / lack of understanding the rules, but who knows.

[Zach Solomon @ZachSB]

@dneighbors Agreed. I keep finances local - I also don’t really need frontier LLMs for that.

[Zach Solomon @ZachSB]

@SawxSouth I don’t get why they even released it

[Zach Solomon @ZachSB]

@RaminNasibov Design so good the end product feels like a no brainer. 👏

[Zach Solomon @ZachSB]

@GustavS @danielcberk @Spotify Doesn’t work when using devices like Google Home, which is a lot of our listening

[Zach Solomon @ZachSB]

@karpathy I see a lot of this in the comms world too -- two worlds: one that's focused on "write me a draft" and another that has built out the comms system with context, architecture, knowledge/memory, voice profiles, and more. Same tool but completely different maturities.

[Zach Solomon @ZachSB]

@VibeMarketer_ the whole thing is different. Like, that completely transforms the sales function, and it’s moving incredibly fast. Really exciting for those who are leaning in and figuring this all out.

[Zach Solomon @ZachSB]

@GamewithDave Roller coaster tycoon 3

[Zach Solomon @ZachSB]

@sophiaamoruso @rrhoover we use notion, but it will be part of a Hermes/Open Claw solution soon. I have manuals, projects, and other resources in one project. Eventually, a local llm will have all spending/investments, travel, home projects, health and so on in one place, continuously improving.

[Zach Solomon @ZachSB]

@DennisonBertram @sophiaamoruso Interested to see what this looks like

[Zach Solomon @ZachSB]

@mattlindner @chi_urbanist I don’t get what was behind the random day they painted half that building yellow

[Zach Solomon @ZachSB]

I've been doing exactly this for comms strategy work — a structured markdown workspace with 100+ curated documents, voice profiles, context layers, and templates that the AI operates as a teammate. Organizing the knowledge and context is the differentiator.

[Andrej Karpathy @karpathy]

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.

[Zach Solomon @ZachSB]

@landon20s Chicago sports legends. Jordan, Payton, Banks, Hull, Thomas, Butkus, Ditka. But you need more meeting space 😬

[Zach Solomon @ZachSB]

when starting my career in comms, it was almost frowned upon to be technical — they positioned you as an individual contributor doer, not a strategic high potential employee. Unreal how wrong they were, and that never sat with me well. AI has allowed me to merge both.

[shirish @shiri_shh]

generalists are about to win big

If you understand a little of tech, business, and people, and can connect everything fast.

you're sitting on a goldmine right now.