Field reports from the TechKis team.
Engineering notes, architecture deep-dives and honest takes on what we’re learning while shipping AI-first software.
Top highlights
- 9 min read·AI EngineeringCostEngineering Practice
What an AI pilot actually costs — a line-item breakdown
Most AI pilot budgets are wrong because they only price the build. Here's the honest line-item breakdown — discovery, data prep, evaluation, build, inference and the run cost nobody quotes — and where pilots quietly overrun.
Read post - 9 min read·PerformanceCore Web VitalsSEO
A Lighthouse 100 doesn't mean your site is fast
Lighthouse is a lab simulation on a throttled mid-range phone. Google ranks on field CrUX p75 for LCP, INP and CLS. Here's why the two disagree, and what to measure instead.
Read post - 9 min read·AI EngineeringSEOContent
Getting cited by AI search — how to be the answer, not the tenth blue link
ChatGPT, Perplexity and Google AI Overviews cite sources instead of listing links. Here's how answer engines pick what to quote, and how to be the citation.
Read post
All posts
29 posts · page 5 of 5
- 10 min read·RAGVector Searchpgvector
pgvector vs Pinecone vs Qdrant at 10M chunks — what we'd actually run
A practical comparison of pgvector, Pinecone and Qdrant for RAG at the 10-million-chunk threshold — where Postgres still wins, where managed beats self-hosted, and the real cost trade-offs that actually decide it.
Read post - 9 min read·DesktopTauriElectron
Tauri vs Electron in 2026 — binary size, RAM, auto-update and the parts no one benchmarks
A practical comparison of Tauri and Electron for desktop apps in 2026 — measured binary size and RAM, the real auto-update stories, code-signing pain, and where each one earns its keep.
Read post - 8 min read·Google I/OGeminiAI Engineering
Google I/O 2026 — what stood out to our team
Our take on the Google I/O 2026 announcements that actually matter for shipping AI-first software in production — Gemini, Android, web platform and developer tooling.
Read post - 9 min read·MobileReact NativeExpo
React Native vs Expo vs Flutter for AI apps in 2026
An opinionated comparison of React Native, Expo and Flutter through the lens of shipping AI-first mobile apps — streaming UIs, on-device inference, native bridges and the team-velocity trade-offs that actually decide it.
Read post - 9 min read·AWSBedrockLLM Infrastructure
AWS Bedrock vs direct provider APIs — cost, lock-in and the procurement reality
When to call OpenAI and Anthropic directly versus going through AWS Bedrock — a practical breakdown of cost, latency, model availability, governance and the procurement reality that actually decides the answer on most enterprise projects.
Read post