/// BLOG

LOG: BLOG

Ruminations on code, intelligence, and the architectural voids of the modern web.

2026-10-15

Ecommerce Faceted Navigation and SEO - Which Filters to Index and Which to Block (noindex, canonical, robots.txt)

Which ecommerce filters to index and which to block: a rule with thresholds, noindex vs canonical vs robots.txt vs nofollow in one table, a check of filters in 11 Polish stores, Shopify, WooCommerce, PrestaShop and IdoSell defaults, URL counts at two scales, robots.txt rules, a script and two pitfalls.

15 min

Ecommerce Keyword Research - How to Map Queries to Categories, Filters and Products

2026-10-12

AI & SEO

Ecommerce Keyword Research - How to Map Queries to Categories, Filters and Products

How to do ecommerce keyword research and map queries to categories, filters, product pages and guides: a mapping rule with thresholds, a keyword map from 1,109 Google suggestions and real store product counts, Search Console, Keyword Planner, Senuto and Ahrefs with prices, a sheet template and a cannibalisation script.

14 min

Ecommerce SEO Strategy - a 3, 6 and 12-Month Roadmap

2026-10-11

AI & SEO

Ecommerce SEO Strategy - a 3, 6 and 12-Month Roadmap

How to build an ecommerce SEO strategy ranked by revenue: a one-day diagnosis from Search Console, a priority matrix, a 3, 6 and 12-month roadmap with KPIs, SEO vs Google Ads cost at two scales, an audit of product data in 26 Polish online stores and a quick-wins script.

15 min

Polish Embedding Models - bge-m3, multilingual-e5 or PolDense? A Test on 900 Questions

2026-10-11

AI & Automation

Polish Embedding Models - bge-m3, multilingual-e5 or PolDense? A Test on 900 Questions

Which embedding model should you use for Polish RAG? I tested 10 models and BM25 on 900 real customer questions: results, robustness to missing diacritics and typos, the prefix that costs 30 points, the PIRB ranking, API prices, cost at two scales, licences and code.

14 min

AI Vendor Lock-in - How Not to Depend on One Model Provider (Multi-Model Strategy and Plan B)

2026-10-10

AI & Security

AI Vendor Lock-in - How Not to Depend on One Model Provider (Multi-Model Strategy and Plan B)

Every AI model has an expiry date: recently retired Anthropic models lived 12-16 months, and companies got about 60 days to migrate. I show where provider lock-in lives, how to build an access layer for several models, how tests cut a model switch down to one day, plus a 10-question test.

13 min

AI Agent Security - Permissions, Sandboxing and an Audit Trail Before an Agent Breaks Something

2026-10-06

AI & Security

AI Agent Security - Permissions, Sandboxing and an Audit Trail Before an Agent Breaks Something

53% of companies say their AI agents have exceeded their intended permissions. What the Replit, GitHub MCP and EchoLeak incidents teach us, the lethal trifecta and the rule of two, and five layers of protection: permissions, sandbox, limits, human approval and an audit trail.

13 min

AI Adoption in Your Team - Why Licences Sit Unused and How to Get People Using AI

2026-10-05

AI & Automation

AI Adoption in Your Team - Why Licences Sit Unused and How to Get People Using AI

According to Gartner, only 5% of companies that finished a Copilot pilot planned a larger rollout, and just 8% of employees use AI daily. Five kinds of resistance and what works, the role of managers, AI champions, measuring adoption at three levels and a 90-day plan.

11 min

How to Choose an AI Implementation Partner - 15 Questions, Warning Signs and Contract Clauses

2026-10-05

AI & Automation

How to Choose an AI Implementation Partner - 15 Questions, Warning Signs and Contract Clauses

According to MIT NANDA, 95% of AI pilots show no financial impact, yet deployments with partners succeed in 67% of cases. In-house, partner or hybrid, 15 questions to ask, warning signs, contract clauses and a pilot with success criteria.

11 min

TOTAL_RECORDS: 0110

PAGE 1/14

SYSTEM_STABLE