https://alangoh.comPractical AI training in Malaysia for owners, leaders and teams. Build repeatable AI workflows around real work, with method and guardrails that hold up after the demo.https://alangoh.com/notes/three-ways-to-get-ai-work-done/https://alangoh.com/notes/three-ways-to-get-ai-work-done/Mon, 19 Oct 2026 09:00:00 +0800Your team can learn to do the AI work, build it with a guide alongside, or have it built and then own it. Three questions tell you which way fits each piece of work, and most businesses use all three.https://alangoh.com/notes/fix-it-before-you-automate-it/https://alangoh.com/notes/fix-it-before-you-automate-it/Mon, 12 Oct 2026 09:00:00 +0800Automation runs whatever version of the task it is given. A quick test tells you whether a task is ready — and getting it ready is usually smaller work than people expect.https://alangoh.com/notes/it-guessed-because-you-didnt-say/https://alangoh.com/notes/it-guessed-because-you-didnt-say/Fri, 02 Oct 2026 09:00:00 +0800When AI gives you a generic answer, it has usually filled the details you left out with its best guess. Telling it the details you care about changes the answer.https://alangoh.com/notes/when-ai-tools-change-skills-stay/https://alangoh.com/notes/when-ai-tools-change-skills-stay/Sat, 26 Sep 2026 09:00:00 +0800AI tools keep changing. The skills a team builds, such as breaking work into steps, briefing AI clearly and checking what comes back, carry over to every new tool, which makes them the investment that lasts.https://alangoh.com/notes/hand-ai-the-boring-task-first/https://alangoh.com/notes/hand-ai-the-boring-task-first/Sat, 19 Sep 2026 09:00:00 +0800The best first task for AI is a routine one: it comes back every week, you could write its steps down, and you can check the answer. The impressive one can come later.https://alangoh.com/notes/five-skills-for-good-ai-results/https://alangoh.com/notes/five-skills-for-good-ai-results/Wed, 09 Sep 2026 09:00:00 +0800Getting good results from AI comes down to five skills: break the task down, give it the context, set the limits, refine the draft, and judge the result. When an answer disappoints, the missing skill tells you what to fix.https://alangoh.com/notes/keep-your-ai-instructions-in-a-file/https://alangoh.com/notes/keep-your-ai-instructions-in-a-file/Wed, 02 Sep 2026 09:00:00 +0800The instructions you give AI for a repeated task are worth keeping in a plain document. Use the file every time, update it whenever you correct a result, and the task gets better with every run, whoever on the team runs it.https://alangoh.com/notes/split-asking-doing-and-checking/https://alangoh.com/notes/split-asking-doing-and-checking/Wed, 19 Aug 2026 09:00:00 +0800AI work goes better with three roles kept apart: you decide and write down what good looks like, a fresh AI session does the work from that instruction alone, and someone with no stake in the result checks it against your description.https://alangoh.com/notes/ai-can-be-confidently-wrong/https://alangoh.com/notes/ai-can-be-confidently-wrong/Wed, 05 Aug 2026 09:00:00 +0800AI writes by predicting what fits, so it can give an answer that is specific, fluent and wrong. Treat every AI answer as a draft, and check the parts that carry weight, such as numbers, references and names, before you rely on it.https://alangoh.com/notes/two-kinds-of-automation/https://alangoh.com/notes/two-kinds-of-automation/Wed, 22 Jul 2026 09:00:00 +0800AI automation handles work that varies. Coded automation handles work that must run the same way every time. Most good workflows use both, each where it fits.https://alangoh.com/notes/brief-ai-like-a-person/https://alangoh.com/notes/brief-ai-like-a-person/Wed, 15 Jul 2026 09:00:00 +0800Search engines taught us to type a few keywords. AI works from meaning, so it gives better answers when you describe the situation and what the result is for, in plain sentences, the way you would brief a person.https://alangoh.com/notes/build-what-only-you-can-build/https://alangoh.com/notes/build-what-only-you-can-build/Wed, 01 Jul 2026 09:00:00 +0800AI now lets almost anyone build software, senior leaders included. For a leader, the most valuable hours usually still go into direction: deciding what should be built, and why.https://alangoh.com/notes/three-permissions-your-team-needs/https://alangoh.com/notes/three-permissions-your-team-needs/Wed, 24 Jun 2026 09:00:00 +0800Teams take up AI when they feel free to use it the way it works best: to ask basic questions, to hand it the effort, and to keep asking for a better answer. A manager can grant all three in one sentence.https://alangoh.com/notes/the-cost-is-keeping-it-alive/https://alangoh.com/notes/the-cost-is-keeping-it-alive/Wed, 10 Jun 2026 09:00:00 +0800An automation keeps paying off for as long as someone looks after it. Decide who that is, and when they'll do it, before you build.