Tech.AI.Fin

Hey business owners! Want a smarter, hassle-free way to track and forecast your cash flow? Meet your AI-powered cash flow agent at cashflow-ai.com. Register for free today.

Finished The Little Book of Common Sense Investing by John Bogle. My biggest takeaway: Owning the market efficiently: Low-cost index funds, disciplined asset allocation, and keeping fees rock bottom still look like the most durable edge for long-term investors.

Pleasantly surprised to find that Claude Desktop now has a GitHub-style overview feature! 🤯 Love the way to code like a pro🚀 #ClaudeAI #Anthropic #Productivity

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.

First time trying transferred my first domain to new host provider. @Cloudflare! 🚀 Saving money on renewals means more money for coffee. ☕️ New experience unlocked! 🔓✨ #Cloudflare #WebDev #TechTips

AI has erased the barrier of coding. You can now build business apps with ideas, not syntax. Just as Excel revolutionized PCs decades ago, AI empowers domain experts to turn imagination into enterprise reality. A new era of productivity has begun. 🚀

My server was DDoSed and injected. On the bright side, I finally have the opportunity to handle network threats...

Lying on the haystack of the old IT world, witnessing the rise of the new AI era.

My Snowflake learning has been delayed by my recent job. Every time I start working on it, it brings new experiences to me. Today, I am connecting Snowflake from VS Code IDE. #snowflake #datawarehouse

Things that happened this year: 1. Built AI apps 2. Provided consulting services 3. Participated in the biggest deal in career history 4. Went through death threat and broken body 5. Quit a toxic job 6. Learning new things and facing challenges 7. Swam for 2 months Happy 2026!

Now I'm facing a multi-party business contract negotiation, which is far less severe than sudden death. I was struggling in the middle of it. I believe it will pass anyway; nothing compares to time.

Three years ago, my business was hours away from bankruptcy and a lawsuit. I was short $1 million to pay an early-due bank loan, which the bank indicated it would not renew. We borrowed the money at the last minute, and it took six months to clear the debt. #Entrepreneurship

Red‑eye flight on a budget airline. A middle‑aged man just completed his full wash‑up and skincare routine for the first time ever — in an airport restroom.#midlife

The past two months have been focused on data warehousing and ETL. Doing Snowflake tests to catch up on my learning progress. #snowflake #datawarehouse

🌳 tETH Turns One: A Review In one year, $tETH has grown to 81K+ ETH ($300M) in TVL, 47K+ users, and 2 chains, delivering on its promise of steady, low-volatility yield. But the real story isn’t just returns. It’s how the strategy held strong through DeFi’s “hard mode.” 🧵👇

🌳 TIP 4: TREE Buybacks A new TIP proposal is live to allocate 50% of MEY fees from $tETH toward $TREE buybacks. Acquired TREE will be held in a designated address managed by the DAO Treasury, strengthening long-term alignment between protocol growth and tokenholder value. 🧵👇

🌳 Q3 2025 was Treehouse’s biggest quarter yet. We launched $TREE through Gaia TGE, expanded tAssets across major protocols and chains, and saw real-world adoption of DOR with institutional integrations. Here’s everything we built this quarter. 🧵👇 treehouse.finance

I've been trying crypto futures trading these days, with gains and losses. Every time I bet on a direction, I start to lose money. Today I picked up the book Antifragile, hoping it could give me some hints on this disordered period.

zkPass × @BinanceWallet Booster Program - Phase 1 Goes Live → Starts Nov 5, 12:00 UTC → Eligibility: Users with 61 + Alpha Points → Rewards: 10,000,000 $ZKP → Complete verifiable quests, power your proofs with VOLEitH, and earn what you prove. Proof is the new Alpha. 🌐 ✅

@REVApay_ai send 6 USDC on BNB to @ray_zz_mylife

null(ID) 🟩 Send me any stablecoins at my Pay(ID) by @r3vl_xyz – reveel.id/pay/null 🚀 Get yours now: reveel.id