BLOG · topic
LLM + RAG architecture
Production-ready LLM + RAG systems · what makes them faster, cheaper, and more accurate than a demo.
posts37projects20case studies03
Notes from the studio.37 / posts
- 0130 Sept 2026Q3 2026 roundup: what shifted, what we shipped, what broke→
- 0229 Aug 2026Does your AI phone assistant have to tell callers it's AI? (EU, 2026)→
- 0328 Aug 2026AI Receptionist for Auto Shops: Never Miss a Booking Call (2026)→
- 0428 Aug 2026AI Receptionist for Clinics: End Missed Calls and No-Shows (2026)→
- 0527 Aug 2026AI Voice Receptionist for Business: Costs & Build vs Buy (2026)→
- 0601 Jul 2026Q2 2026 roundup: what shifted, what we shipped, what broke→
- 0702 Jun 2026n8n vs Make vs custom code: 2026 automation stack→
- 0802 Jun 2026AI agent pricing 2026: what an autonomous agent costs→
- 0902 Jun 2026H1 2026 in review: what changed for EU software teams→
- 1002 Jun 2026AI in logistics & supply chain: 2026 SME guide→
- 1114 May 2026The EU AI Act in practice: a 2026 guide for AI teams→
- 1214 May 2026Self-hosted AI or the API? When to run your own LLM in 2026→
- 1309 May 2026EU AI Act for Hungarian startups: 2026 founder's guide→
- 1409 May 2026Hiring an AI Development Team in Budapest · 2026 Guide→
- 1506 May 2026Direct Booking vs OTAs in 2026 · the Math and the Tech→
- 1606 May 2026Proptech stack for EU builders in 2026: choices that scale→
- 1706 May 2026LegalTech Due Diligence 2026: What to Check→
- 1806 May 2026Manufacturing AI in 2026 · what works on the floor→
- 1929 Apr 2026How to Hire an AI Development Team · 9 Questions for 2026→
- 2029 Apr 2026How Much Does AI Development Cost in 2026? EU Rates→
- 2126 Apr 2026RAG's three failure modes (and the diagnostic table)→
- 2226 Apr 2026Build an LLM Eval Harness in 200 Lines of TS→
- 2326 Apr 2026Why your AI agent leaks money: 6 prompt-cache wins→
- 2426 Apr 2026OWASP LLM Top 10 v2 · what changed and what to ship→
- 2523 Apr 2026On-device LLMs in 2026: Gemini Nano vs Apple Intelligence→
- 2622 Apr 2026pgvector at 10M+ rows: index, queries, real numbers→
- 2722 Apr 2026LLM prompt caching in production · a 60-80% cost cut→
- 2822 Apr 2026Agentic AI · the safe tool-use pattern we ship by default→
- 2922 Apr 2026LLM evals-as-code · the CI gate we run on every RAG deploy→
- 3020 Apr 2026What an AI security audit actually checks in 2026→
- 3120 Apr 2026How to ship a production AI chatbot in 14 days→
- 3218 Apr 2026LLM prompt injection playbook · the 2026 attack surface→
- 3314 Apr 2026MCP (Model Context Protocol): what it means for LLM agents→
- 3408 Apr 2026Shipping AI agents that actually work in production→
- 3505 Mar 2026GDPR + AI: training on user data 2026 · what's allowed→
- 3618 Feb 2026EU AI Act for SaaS: what you actually have to do in 2026→
- 3722 Jan 2026Picking a vector DB in 2026: pgvector, Pinecone, Weaviate→
SHIPPED WORK20 / shipped
- 012026Vilya Protection→
- 022026Methora→
- 032026AutoImport→
- 042026ClarixAI→
- 052026AIHealthIQ→
- 0620263D AI Property→
- 072026AxisFit→
- 082026autoszoftver.hu→
- 092026SimulSpeak→
- 102026JuriSafe→
- 112026Use AI Easily→
- 122026AI Chatbot Maker→
- 132026PhisGuard→
- 142026MCP Security Layer→
- 152026n8n AI Workflow Generator→
- 162026The Truth AI News→
- 172026Multi-Agent Crypto Trading→
- 182026BetEdge→
- 192025DField Poker→
- 202026GlowUp→
CASE STUDIES03 / studies
Liked what you saw? Let's build yours.
Short email or a 30-min call · 24h reply.
Start a project