Read-only staging package. Nothing here is connected to live email, CRM, calendar, ads, or spend.
Link Index · H2
H2 — Day 2 homework
High-level Day 2 summary supplied as homework context for the final project.
Core idea: context beats prompting. A mediocre prompt with the right source material is stronger than a polished prompt with no context.
The model behaves like a capable new hire who has no access to the founder's files, history, inbox, or business until it is briefed.
Context-engineering loop: gather real material, clean noise, load it into a persistent Project, then ask short questions.
Useful source types: call transcripts, YouTube or podcast transcripts, Google Drive files, websites, dictation, and email threads.
A Claude Project should contain three layers: short instructions, source knowledge, and two or three examples of the desired output.
Minimum source mix for voice and positioning: one transcript, one careful document, and one public page.
Instructions should state who the model is, who it writes for, voice rules, prohibitions, and the finish line for acceptable output.
Chat is best for one-step answers or drafts. Agentic work is best when the task has dependent steps, touches tools or files, repeats, or needs checking.
Claude Cowork handles multi-step tool work; Claude Code handles files and code with a higher ceiling and steeper setup.
Safety rules: never paste passwords, API keys, bank details, or sensitive material into a public AI tool; verify specific claims before client use.
Markdown is the portable structure shared by Claude, Obsidian, Notion, GitHub, and other tools.
Homework: build a Project for Kasim Aslam from a transcript, document, and public page; test it against real source material.
Relevance to the final project
Build the Pareto Talent research brain from multiple source types before writing the offer, funnel, or ads.
Keep facts traceable and flag unknowns instead of filling gaps with plausible claims.
