How-To tag | eYssen AI

63 Articles

How-To ×

Refactoring Safely with AI

Krisztian Eyssen, Krisztian Eyssen

Refactoring Safely with AI

Roughly **one AI refactoring in twenty changes the behaviour while the tests stay green.** That is not an estimate: it falls out of multiplying two published measurements. And that number is the reason the order of your checks matters more than the prompt here.

Advanced Developer How-To Coding & Development

Sep 13, 2026

Your First Task with a Coding Agent

Krisztian Eyssen, Krisztian Eyssen

Your First Task with a Coding Agent

There is a metric for how long a task an agent solves with a 50% chance of success. For the best model that is roughly **five hours of human work**. Fifty percent means it gets every second one wrong. Do not put your first task there — and give it something it can run itself.

Beginner Developer How-To Coding & Development

Sep 13, 2026

Testing and Monitoring Your Agents

Krisztian Eyssen, Krisztian Eyssen

Testing and Monitoring Your Agents

In a survey of 1,300 practitioners, **39.8%** of respondents ran offline evaluation and **32.5%** ran it online — while everyone named **quality** as the main obstacle to getting into production, more than twice as often as cost or security. Everybody logs. Measuring whether the agent is doing a good job is something most people cannot do. This article is about how you can.

Advanced Developer How-To Agents & Automation

Sep 13, 2026

Scheduled and Background Agents

Krisztian Eyssen, Krisztian Eyssen

Scheduled and Background Agents

One vendor's documentation describes its own interface like this: **"a green status means the session started and exited without an infrastructure error. It does not mean the task succeeded."** That one sentence sums up the entire problem with background agents. And according to one measurement, **more than a third** of failures are exactly this: the agent **does not know when to stop**.

Advanced Business owner How-To Agents & Automation

Sep 13, 2026

Subagents and Parallel Work

Krisztian Eyssen, Krisztian Eyssen

Subagents and Parallel Work

The most quoted number says a multi-agent setup is **90.2% better** than a single-agent one. The same source also writes that it uses **fifteen times as many tokens**, and that **80% of the performance difference is explained by token consumption alone**. And a 2026 paper showed it formally: **at an equal token budget, the single-agent system wins.** The value of splitting a task is not parallelism — it is **clean context**.

Advanced Developer How-To Agents & Automation

Sep 13, 2026

Permissions: The Least-Privilege Agent

Krisztian Eyssen, Krisztian Eyssen

Permissions: The Least-Privilege Agent

Three things are needed for a data leak: **private data**, **untrusted content** and **an outbound channel**. Any two of them on their own are safe — all three together are not. And this is not theory: according to a study of 7,973 live servers, **40.55% of them expose their tools with no authentication**.

Advanced Business owner How-To Agents & Automation

Sep 13, 2026

Human in the Loop: Where to Put the Approval

Krisztian Eyssen, Krisztian Eyssen

Human in the Loop: Where to Put the Approval

The human who approves is not a neutral safety layer. In a 2026 randomised clinical trial, **44 doctors, after twenty hours of AI literacy training**, still performed **14 percentage points worse** whenever the AI handed them a wrong recommendation. Training does not protect against this. What does: **few approval points, but real ones.**

Intermediate Business owner How-To Agents & Automation

Sep 13, 2026

Your First Automation That Is Actually Useful

Krisztian Eyssen, Krisztian Eyssen

Your First Automation That Is Actually Useful

Most companies start where the return is worst. According to a report covering 52 organisations and 153 executives, roughly **70% of the AI budget goes to sales and marketing**, while **back-office processes** delivered faster payback and cleaner cost reduction. And there is one number that rearranges everything: for an automation running once a day, **the platform fee is typically more than a hundred times the model cost.**

Beginner Business owner How-To Agents & Automation

Sep 13, 2026

Project and Task Management with AI

Krisztian Eyssen, Krisztian Eyssen

Project and Task Management with AI

For this area **there is no controlled-group measurement** — and that is worth saying out loud before anyone quotes a number at you. What there is: task-level data from two government rollouts, one set of which its own publisher calls an **upper estimate**; a finding that 60% of participants had to edit the output substantially; and up to 7% for whom the work **grew** because of it.

Intermediate Office worker How-To Work & Productivity

Sep 12, 2026

A Personal Knowledge Base with AI

Krisztian Eyssen, Krisztian Eyssen

A Personal Knowledge Base with AI

Your notes become useful to a machine when **they have structure** — and that is not a matter of taste. In a measurement comparing 36 chunking methods, the split that followed the structure gave **24% top-1 accuracy**, while blind, fixed-length cutting gave **2–3%**. A tenfold difference, out of the same material. A markdown heading is not formatting, it is data structure.

Intermediate Everyone How-To Obsidian Work & Productivity

Sep 12, 2026

Note-Taking That Feeds Your AI

Krisztian Eyssen, Krisztian Eyssen

Note-Taking That Feeds Your AI

The usual advice is to tag your notes. The most thorough measurement says **most enrichment makes things worse**: a reliability score, a contradiction flag and a provenance chain **lowered accuracy on every task tested**. One single piece of metadata helped meaningfully — **the date**. This article is about what is actually worth writing down.

Beginner Everyone How-To Work & Productivity

Sep 12, 2026

Measuring AI's Effect on Your Marketing

Krisztian Eyssen, Krisztian Eyssen

Measuring AI's Effect on Your Marketing

A research team measured how badly people misjudged their own speed-up when they worked with AI. The error averaged **39 percentage points**. That number is bigger than the effect you are trying to measure — which is why you need four metrics, not an impression.

Intermediate Business owner How-To Marketing & SEO

Sep 12, 2026

1
2
3
4
5

Follow Us

Archives

Tags

Audience

Everyone Office worker Marketer Creative Business owner Developer

Category

AI Basics Safety & Risk Regulation & Policy Infrastructure & Chips Business & Deals Prompting Models & Market Writing & Content Images & Design Video & Audio Marketing & SEO Coding & Development Work & Productivity Agents & Automation Data, RAG & Self-Hosted AI Business, Law & Security

Format

How-To Explainer Comparison Tool review Recipe Template Troubleshooting Case study Glossary News

Level

Beginner Intermediate Advanced

Tool

OpenRouter Perplexity HeyGen ElevenLabs Suno Obsidian MCP n8n Ollama EYAS

Others

Odoo