Context Engineering for Teams
AI isn't Excel — learning functions isn't enough. You need to change the way you work.
Context engineering, AI v tíme, PostgreSQL.
AI isn't Excel — learning functions isn't enough. You need to change the way you work.
Enough theory. Concrete examples of how I use Context Engineering in real situations — from debugging to solar panels.
There's currently big hype around Vibe Coding. But is it really enough for production code?
The answer isn't about which tool is 'the best' overall. It's about which is best for a specific thing.
AI shouldn't replace your work. It should help you overcome yourself and push you to a higher league.
Break down problems, not prompts. What I've used at work for 20+ years, I also use with AI.
Dumping everything into the context window is not context engineering. Here's what actually works.
Keď AI nedáva použiteľné výsledky, zvyčajne nechýba lepší prompt, ale kontext. Desať rád z mojej prednášky na CodeCone 2025, ako AI zadávať prácu tak, aby jej rozumela.
Context is not 'more information'. Good context is structured, relevant, and scoped to the task.
Most teams treat AI like a magic box. The real skill is structuring context — not crafting prompts.
78% of e-commerce visitors go to search first. 68% never return after bad results. Here's the math, the four approaches, and how we implemented hybrid search for 11,000+ products in 3 days.