This Course briefly explains what Obsidian Metrics is and how the framework that it's built on works. Full overview of the tiers and benefits.

A taste of what's inside the full Classroom. Premium ($19/mo) unlocks the complete Platform Drop, Systems Breakdown, and Tools libraries.

Where to Park Cash Now — Free PDF Download The Fed cut rates again in December 2025 to a 3.50–3.75% target range, with projections pointing to roughly 3.4% by end of 2026 and 3.1% by end of 2027. HYS rates have already compressed to the 3.5–4.5% range, and most of the headline 4%+ rates are 90-day promotional offers that drop after the intro period. The question every system-builder is now facing is: where does cash go when the easy yield era ends? This is a framework for thinking about it.

Start your journey for investment knowledge here. What we cover here: -investment terms -Strategies -Markets -Platforms -Investment types This is where you want to go if your looking to invest like a pro.

This Classroom module is ONE module containing TWO formats of the same content: 1. The written deep-dive post (The Body Below) 2. The embedded video (The Classroom Premium Video) The written post and the video are NOT separate destinations — they live inside the same Classroom module.

Income platforms generate returns from deployed capital — interest, dividends, distributions, or yield on a regulated principal. Each breakdown covers the platform mechanics, fee structure, historical rate range, counterparty, and where the platform fits inside the 13-system library. Rates are historical, not predictive. Verify on the official site before deploying capital.

Open-access tools for every member of the community. Free to use without an account. Connect them to your Obsidian Metrics login and they pull in the systems you built in the Lab so you can use them across the whole framework.

4 complete procedures you can run start to finish, not tips. Each has a trigger that tells you when to use it, inputs it needs, literal steps in order, a check that can actually fail, where a human looks before anything leaves the building, and a clear definition of done. They cover jobs you already have not invented examples. Work through one and you get a repeatable process not a conversation you have to reinvent each time. Start here if you get decent answers and want them consistently.

5 lessons on getting a genuinely useful answer instead of a plausible one. Why vague requests produce vague output, what context to give and what to leave out, how to push back on a first draft rather than accepting it, and how to tell when the model has understood you. Every lesson has prompts you can paste, a named mistake, and the fix for it. By the last one you will run a full document task from blank page to finished piece. No technical background needed, nothing here requires a paid tool.

4 lessons on doing real work rather than asking questions. Break a multi-step job into pieces the model can actually hold, the frame-then-fill method for marking what you don't know instead of it be invented, when to abandon a conversation and start clean, and how to have the model check its own output in a way that can come back honestly. This is the tier where most people stop improving, usually because they treat the agent as if it remembers everything. Take Beginner first if you are new.

4 lessons on building things that keep working when you aren't watching. Written procedures instead not habits, handoffs you can inspect when something goes wrong, and the rule that governs all of it: a check that cannot fail is not a check. How to catch the model in a lie, which is the failure that costs the most and announces itself the least. This tier assumes you run multi-step work and want reliable rather than impressive. Not a coding course, but you will get better at directing one.

4 lessons on how the pieces fit together: the model, context you give it, tools it can reach, and memory it carries between sessions. Once you name which of the four is failing, most problems stop being mysterious. Includes 6 ways a setup breaks, what each one looks like. An assistant with memory of your situation beats a smarter one starting cold every time, which is why the arrangement matters more than the model. Useful whether you are assembling your own stack or judging another one.

An honest account of what these systems do well, where they fail, and how much to trust them for a given job. The failures are not softened, because knowing where the floor is worth more than a list of features. Covers what is genuinely reliable, what looks reliable and is not, and how to calibrate trust by task rather than by vendor claim. Read this before the training tiers. It is the shortest path to using AI well, and the fastest way to stop being impressed by things that do not deserve it.