Search the Practicum

Type two letters · ↑↓ select · Enter open · Esc close

Type at least two letters.

What to learn now

This is a live Practicum page for Chapter 11. Verified: 2026-08-18. Specific tools and market signals become outdated faster than the book, so check the date. This is a starting point for this quarter, not a permanent list.

Do not try to learn everything. Choose one skill that supports your 90-day plan in the personal 90-day plan, then practice it on real work.

Foundations for everyone

  • Work with AI as a pair: ask for options and check the result instead of accepting a finished answer, as discussed in Chapter 8.
  • Build context instead of collecting prompts: create a personal knowledge and template base that AI can use.
  • Publish one proof: finish one result and show it under your own name, as discussed in Chapter 7.

Skills with growing demand

See skills that pay for current figures. The short version: the market increasingly rewards analytical thinking, AI literacy paired with domain knowledge, sometimes called a T-shaped skill set, agent orchestration, and communication and trust. Demand is falling for routine manual output by itself.

One 2026 finding should change what you practice. Anthropic's usage data shows that experienced users automate work successfully far more often than newcomers, and that collaborative use has overtaken full delegation. The skill that pays is not "can I get the model to produce something." It is knowing which part of a job to hand over and how to check what comes back. Practice that on real work, not on prompt collections.

Skills for your profession

Use your role hub in the catalog, especially the "minimum stack" and "what to try" sections.

Tools

Start with the tool map and simple agents.

Market signal: AI ability is becoming a standard hiring requirement

Employers have started to treat the ability to work with AI as a basic requirement, not an extra benefit.

  • Russia: Sber said in Jul 2025 that employees and candidates must know how to apply AI. In its GenAI 2025 market review, red_mad_robot compared AI ability with knowing how to use Word. Check the exact wording in the primary source.
  • Global: in an Apr 2025 memo, Shopify CEO Tobi Lutke asked teams to show that AI could not do the work before requesting more people or budget. The phrase was "hire an AI before a human." Similar signals from Fiverr and other companies show a direction, not a rule that applies everywhere.
  • In hiring data, not memos: postings requiring experience with AI coding tools grew about 340% between Jan 2025 and Jan 2026, while postings for pure implementation roles fell about 17%. That is one occupation family, but it is the demand curve doing what the memos only talk about. See what the market pays.

Read this without hype. It does not mean that AI has replaced everyone. Management expectations are changing, and value is moving from the volume of output toward outcomes produced with the best available AI setup. There is a counterpoint: Jensen Huang asked companies to require less blind AI use from employees. There is also a darker use of the slogan, when a company calls AI mandatory while cutting staff. For you, the practical conclusion is narrower: basic AI literacy in your profession is becoming a baseline expectation rather than an advantage. Volume 2, Chapter 5, "A Staff Without Hiring," shows what this looks like at company level.

Update rule: review once a quarter. Version: 2026-08-18.

Sources (verified 2026-08-18)

Use reports for salary premiums and demand, not intuition. Check Russian figures against the primary source.

What to learn now