$ ls ./blog
Articles and hands-on tutorials about developer tools, LLM engineering and building in public.
AI Engineering ★ Featured Tutorial
Trailing commas, missing brackets, wrong fields? Here is how prompting, JSON mode, structured outputs, schema validation and automatic repair fit together — from least to most reliable.
· 2 min read
How input, output and cached tokens are billed, a worked monthly estimate for a support bot, and how much prompt caching, smaller models and output limits actually save.
#llm#cost#prompt-caching#api
Developers paste tokens, logs and customer data into online tools every day. The reasoning behind “your data never leaves the browser”, how it’s built, and what it costs.
#build-in-public#privacy#astro
What MCP is, how it differs from function calling, how to configure local and remote servers, and how to fix the five most common errors.
#mcp#claude#cursor#agent
How tokenizers work, how tokens relate to words, why non-English text often uses more tokens, what counts toward the context window, and practical ways to cut token usage.
#llm#token#tokenizer
The curl example from the API docs works, but your fetch call fails? A walk through the differences in method, Content-Type, query strings, auth, redirects and CORS.
#curl#fetch#http#javascript
Chunking is the first gate of RAG quality. Concrete examples of chunk size, overlap, structure-aware splitting and metadata — and how to validate your choices with data.
#rag#embedding#llm
The trade-offs of auto-increment IDs, UUID v4 and UUID v7 — why random UUIDs slow down inserts, what v7 fixes, and the storage details that matter.
#database#uuid#postgresql#mysql
What the five fields mean, a quick-reference table of common schedules, the special day-of-month / day-of-week rule, and how Linux, Kubernetes, GitHub Actions and Spring differ.
#cron#devops#linux
