Od ideje do AI proizvoda u produkciji: realan put
Većina AI proizvoda zaglavi na stvarima oko modela: šema baze, autentikacija, tajne i deploy. Iz iskustva sa Next.js, Supabase, PostgreSQL i Docker
Tag archive
Većina AI proizvoda zaglavi na stvarima oko modela: šema baze, autentikacija, tajne i deploy. Iz iskustva sa Next.js, Supabase, PostgreSQL i Docker
Most LLM apps fail in production because nobody defined what "working" meant, not because the model is bad. I lay out the checklist I use, covering
Most "AI consultants" ship demos, not systems that survive a 3 a.m. incident with retries, evals, cost caps, and observability. I help founders and teams
Hiring an AI implementation consultant means finding someone who owns the full stack, evals, numbers, and handover, not a prototyper who ships demos
Proximity is usually the wrong first filter for AI automation work, most of it lives in APIs, cloud consoles, and dashboards that don't get faster because
We run a sweep. Every cycle, a job walks through a queue of documents, hits the GPU, pulls context,...
Most of my AI integration work is data plumbing, not prompting: I map where data actually lives, find the ugliest system (usually the real source of
You build a RAG pipeline. You embed your documents, embed the query, and reach for cosine similarity...
You build a RAG pipeline. The retrieval mostly works. Queries come back with documents that are in...

Originally published at norvik.tech Introduction Explore how to optimize response times...
I've refined a five-phase AI implementation process, scoping, POC, integration, evaluation, operations, where each phase has a hard exit criterion that
What We Set Out to Build In early 2026, we decided to run an experiment: build a fully...