/ Blog · Page 2

Notes on learning to use this stuff.

Prompting, verification, automation, and how skills actually stick. Written by the people building the courses.

RSS feed
8 October 2026 · 7 min

Real AI agents examples — a coding agent, a support agent, an audit agent, a research agent — what each one's tool-call loop actually does, and where it fails.

7 October 2026 · 6 min

Nano Banana 2.1 launched on 6 October 2026. Prices per image against Nano Banana 2, what changed, and where the saving is smaller than the headline.

7 October 2026 · 6 min

An online Excel viewer opens a workbook in your browser with no install. The real options, what breaks in the move, and where AI actually helps.

7 October 2026 · 5 min

How to calculate a dollars total in an Excel column: AutoSum, the SUM formula, currency formatting that will not break it, and the checks to run first.

7 October 2026 · 5 min

An Excel Table, made with Ctrl+T, auto-expands and uses structured references instead of cell ranges. How it differs from a plain range, worked through.

6 October 2026 · 6 min

Mistral Large 4 launched on 6 October 2026 as a public preview. What Mistral documents on data and training, and what it leaves open.

6 October 2026 · 7 min

Fin ai tools promise instant analysis. A worked example sorting a month of transactions, where the AI slips, and the checks worth running first.

6 October 2026 · 7 min

Excel advanced functions like XLOOKUP, INDEX/MATCH and FILTER, where AI actually helps write one, a worked example, and the checks before you trust it.

6 October 2026 · 7 min

Agentic workflows let an AI tool take several steps on its own. What that actually looks like for one person, a worked example, and what to check.

5 October 2026 · 7 min

Intelligent process automation now means fixed-rule software plus a model reading unstructured input. A worked accounts-payable example, and the checks needed.

5 October 2026 · 7 min

Best practices for campaign automation that actually hold once AI drafts the copy: trigger rules, personalization fields, and a worked abandonment sequence.

5 October 2026 · 7 min

Tools to automate sales workflow mostly automate stage rules already. A worked quote-drafting example shows where AI adds a step, and where it should not.

2 October 2026 · 7 min

Use an AI tool to draft an Excel inventory sheet from a messy list — a worked example, what it gets wrong, and the checks before you trust it.

2 October 2026 · 7 min

Automation software now means two layers since AI arrived. What changed, a worked test for whether a task is ready, and the checks before you trust it.

2 October 2026 · 7 min

Social media automation already handles scheduling. A worked example of drafting captions from real content, what goes wrong, and checks before you queue them.

1 October 2026 · 7 min

Gemini 4 Argon is Google’s new frontier model. See who can access it, announced API prices, the output limit and a practical way to assess its work.

1 October 2026 · 7 min

Gemini 4 Argon was announced on 30 September 2026 but most people cannot use it yet. What Google says about its safety, and what decides your data.

1 October 2026 · 7 min

An Excel alternative search usually hides one of four problems — cost, collaboration, structure, or an AI layer. Match the real one, worked through an example.

1 October 2026 · 7 min

An AI planner turns a task list into a day-by-day schedule fast. A worked example, where it quietly overbooks you, and the checks that actually matter.

1 October 2026 · 8 min

Agentic AI vs generative AI, in practical terms: one drafts and hands back, the other acts across steps unsupervised. A worked comparison and the real risk.

30 September 2026 · 7 min

Manus Cue vs Manus 2.0: compare personal agent identity with an editable project workspace, early access, usage costs and a workshop-planning example.

30 September 2026 · 6 min

OpenAI dots vs Copilot Cowork: compare workspace access, shared Office files, approvals and costs before a practical customer-review trial.

30 September 2026 · 6 min

OpenAI dots vs Perplexity Computer: compare research, continuing work, plans and source checking with a supplier-shortlist trial and review criteria.

30 September 2026 · 6 min

OpenAI dots vs Claude Cowork: compare continuing work, cloud and local access, paid plans and review, then try a bounded operations-report assignment.

30 September 2026 · 6 min

Compare OpenAI dots vs Meta Muse using current official access, connections and approval guidance, plus a practical way to assess ongoing work.

30 September 2026 · 6 min

Compare OpenAI dots vs Manus Cue on identity, invitations, connections, budgets and approvals, with a practical workshop-planning trial you can review.

30 September 2026 · 7 min

Understand OpenAI Dots permissions, connected apps, custom rules and memory. Learn how to review ongoing work and stop tasks without guessing the scope.

30 September 2026 · 7 min

Learn how to use OpenAI Dots for a first work task: desktop setup, a clear brief, source checks, activity review and a recurring task you can control.

30 September 2026 · 6 min

What is OpenAI Dots? Understand the agent, its launch access, cloud work and limits, then decide which ongoing responsibility is worth handing over.

30 September 2026 · 6 min

The Super Intelligence executive order explained: what Trump’s AI-to-SI order requires of federal agencies, what it leaves alone, and the 60-day deadline.