I found DuckDB significantly faster than Pandas and analyzed Claude 3's cost-effectiveness. I also learned about the Tavily search API, Vertex AI’s Model Garden, and how the isTrusted property prevents scraping on sites like Oracle Service Cloud.
There are sites you TRULY cannot scrape even in the browser because of the isTrusted read-only property of events that you can never set to true. Oracle Service Cloud checks for isTrusted in mouse actions.
I investigated Amazon Nova model costs, surveyed why employees avoid internal LLMs, and evaluated JavaScript text splitters. I also learned about Unicode characters in ChatGPT citations and how to install Docker on Windows without admin rights.
I learned teaching techniques from IIT Madras, automated video highlights with OpusClip, and summarized DuckCon #6. I also found that DeepSeek R1 excels at text-to-CAD and gathered OpenAI’s latest prompting advice for reasoning models.
I explored DeepSeek R1 training, how AI models are absorbing app capabilities, and fixing Windows symlinks for Hugging Face. I also discovered DuckDB's built-in notebook UI, Gemini’s YouTube API, and Karpathy-inspired note-taking workflows.
I explored OpenAI reasoning models, Promptfoo for evals, and terminal tools like cmdg and gcalcli. I also learned about Python’s new t-strings, optimized my fish shell startup, and tested the Unsure Calculator for modeling range-based estimates.
I explored low-cost robotics like SO-ARM100, optimized Docker images with multi-stage builds, and adopted Marp for Markdown slides. I also investigated Model Context Protocol (MCP) workflows and techniques for serving structured content to LLM agents.