AI
Every post tagged "AI" · articles, case studies, guides.
- 01→
Q3 2026 roundup: what shifted, what we shipped, what broke
Three months in. SZEP 2.0 live, NAV v3 cutover, AI Act enforcement, OWASP LLM Top 10 v2. Hard numbers, one strong opinion on the consulting tier.
AI solutions · Websites, web apps & online shops · Cybersecurity · Custom software · everything else - 02→
Q2 2026 roundup: what shifted, what we shipped, what broke
Four months in. Eleven shipped projects, real before/after numbers, one strong opinion on what the consulting tier got wrong this quarter.
AI solutions · Websites, web apps & online shops · Custom software · everything else - 03→
AI agent pricing 2026: what an autonomous agent costs
An AI agent is not a chatbot with extra steps - it takes actions, and that changes the bill. Here are the real 2026 ranges and what drives them.
AI solutions - 04→
H1 2026 in review: what changed for EU software teams
Four shifts defined the first half of 2026 for EU software teams. Here's what changed, what it means, and what to do about each one.
AI solutions · Cybersecurity · Custom software · everything else - 05→
AI in logistics & supply chain: 2026 SME guide
Logistics is one of the few fields where AI pays for itself fast - if you point it at the right problem. Here's where it works for Hungarian operators in 2026.
AI solutions · Custom software · everything else - 06→
The EU AI Act in practice: a 2026 guide for AI teams
The EU AI Act is phasing in, it reaches companies far outside the EU, and most of the work is engineering, not legal. Here's how a build team should actually think about it in 2026.
AI solutions - 07→
Self-hosted AI or the API? When to run your own LLM in 2026
Calling the OpenAI or Anthropic API is the right default for most AI features. But data sensitivity, steady high volume, or strict EU residency can flip the answer. Here's the honest decision.
AI solutions - 08→
Hiring an AI Development Team in Budapest · 2026 Guide
What does an AI engagement actually cost in Budapest right now, and how do you tell a serious studio from a prompt-pretender?
AI solutions · Custom software · everything else - 09→
LegalTech Due Diligence 2026: What to Check
Legal-tech vendors all say 'privilege-aware'. Here's the 12-point due-diligence checklist we run before our law-firm clients sign.
AI solutions · Cybersecurity - 10→
Manufacturing AI in 2026 · what works on the floor
Most manufacturing AI fails at the demo-to-production gap. Here's the short list of what actually pays back · and what to skip.
AI solutions · Custom software · everything else - 11→
How to Hire an AI Development Team · 9 Questions for 2026
Most AI agency pitches sound the same. Nine questions surface the difference between a production engineering team and a deck-driven sales motion · ask all nine before signing.
AI solutions - 12→
How Much Does AI Development Cost in 2026? EU Rates
European AI engineering rates in 2026 don't fit on one page · here's the real breakdown by tier, with what's included, what's a red flag, and where the EU compliance overhead actually shows up.
AI solutions - 13→
RAG's three failure modes (and the diagnostic table)
Three failure modes, one table. 30 minutes of diagnosis, then you know what to fix. Stop guessing.
AI solutions - 14→
Why your AI agent leaks money: 6 prompt-cache wins
Six prompt-cache patterns. Real before/after numbers. Most agents leave 60-80% on the table. Fix it this week.
AI solutions - 15→
On-device LLMs in 2026: Gemini Nano vs Apple Intelligence
On-device LLMs are finally usable for production features. Where Gemini Nano wins, where Apple Intelligence wins, and the Hungarian-language gap.
AI solutions · Mobile apps · iPhone & Android - 16→
pgvector at 10M+ rows: index, queries, real numbers
pgvector at 10M rows is not scary · if you pick the right index. HNSW vs IVFFlat, filter patterns, real numbers.
AI solutions · Websites, web apps & online shops - 17→
LLM prompt caching in production · a 60-80% cost cut
Prompt caching is the single biggest LLM cost lever in 2026. 4 patterns, real savings numbers, 2 gotchas worth knowing.
AI solutions - 18→
Agentic AI · the safe tool-use pattern we ship by default
Agentic AI that can send email and move money is not just a chatbot. Here's the safe tool-use pattern we ship.
AI solutions · Cybersecurity - 19→
LLM evals-as-code · the CI gate we run on every RAG deploy
An eval that's not in CI is not an eval. Here's the evals-as-code workflow we run on every RAG project.
AI solutions - 20→
What an AI security audit actually checks in 2026
AI security isn't a checkbox. Here's the nine-point audit we run on every LLM system we ship, plus which bugs turn up most often on systems we didn't build.
AI solutions · Cybersecurity - 21→
How to ship a production AI chatbot in 14 days
Fourteen days from zero to a live AI chatbot your company can actually use. The schedule we follow on every client project, down to what happens on each day.
AI solutions - 22→
LLM prompt injection playbook · the 2026 attack surface
The prompt injection surface is not a single bug · it's five categories, each with a distinct defence. Here's our playbook.
AI solutions · Cybersecurity - 23→
MCP (Model Context Protocol): what it means for LLM agents
MCP is the most important agent standard of the past year. What it means in practice, where we use it, and why to bet on it in 2026.
AI solutions - 24→
Shipping AI agents that actually work in production
From demo to live system: the retrieval, eval, guardrails and cost control we run on every AI project we ship.
AI solutions - 25→
Picking a vector DB in 2026: pgvector, Pinecone, Weaviate
Three serious vector DBs, three very different DNA. Here's the decision framework that held up across our 2026 projects.
AI solutions · Websites, web apps & online shops
Liked what you saw? Let's build yours.
Short email or a 30-min call · 24h reply.
Start a project