AI's breakneck pace of evolution
The list below reflects topics that are either in progress or already written and waiting to be published. The AI space moves fast — what's "upcoming" today may already be "old news" by next week.
-
Claude Usage Dashboard v2 — The current dashboard already tracks session, weekly, API budget, routines, and design usage via a live MCP fetch. The next version will add an 80% session alert: a browser popup that fires when your current session allowance is running low, giving you time to wrap up before Claude cuts you off.
-
Anthropic's MCP Server ✓ — A revolution's revolution. The Model Context Protocol is rapidly becoming the universal connector between AI agents and the tools they use.
-
Sonnet 3.7 is king — A hands-on look at Claude 3.7 Sonnet's reasoning capabilities and why it's leading the pack in several benchmark categories.
-
Vercel V0 + Sonnet 3.7 — Using V0 as a front-end code generation tool paired with Claude's reasoning engine.
-
RAG vs. fine-tuning — When do you retrieve, and when do you retrain? A practical comparison for real-world use cases.
-
Multi-agent RAG — Taking RAG further by orchestrating multiple agents, each with its own specialized knowledge base.
-
AI agents / bots — The continuum from simple rule-based bots to fully autonomous agents — and what sits in between.
-
Computer use with Anthropic — Anthropic's own take on the Computer Using Agent paradigm, and how it compares to OpenAI's Operator.
-
crewAI (multi-agents) — A deep dive into the crewAI framework for building and orchestrating teams of AI agents that collaborate on complex tasks.
Let's connect
Have a topic you'd like me to cover, or want to share your own AI experiments?
