Recent Posts

• An Agentic Case Study: The Messy Work of Building a Gemini Agent to Play Pokémon
  by dbreunigon 6/17/2025, 8:04:09 PM with 0 comments
• The drawbridges come up: the dream of a interconnected context ecosystem is over
  by dbreunigon 6/17/2025, 12:13:00 AM with 14 comments
• Let the Model Write the Prompt: Using DSPy to Decouple Your Task from the LLM
  by dbreunigon 6/15/2025, 7:37:33 PM with 0 comments
• Comparing Claude System Prompts Reveal Anthropic's Priorities
  by dbreunigon 6/4/2025, 9:38:27 PM with 10 comments
• Using 'Slop Forensics' to Determine Model Lineage
  by dbreunigon 5/30/2025, 5:18:02 PM with 0 comments
• Chatbots Trigger Guardrails Differently Based on Your Age, Gender, and Fandom
  by dbreunigon 5/21/2025, 8:16:49 PM with 1 comment
• Breaking Down Claude's 26k+ Token System Prompt
  by dbreunigon 5/8/2025, 10:17:16 PM with 1 comment
• DuckDB is probably the most important geospatial software of the last decade
  by dbreunigon 5/3/2025, 7:30:38 PM with 30 comments
• Explaining Reasoning Models: How they work, how they're built, & why they matter
  by dbreunigon 4/18/2025, 8:17:15 PM with 0 comments
• AI-Powered Apps Are Changing the Dynamic Between Domain Experts and Developers
  by dbreunigon 4/10/2025, 2:20:54 PM with 0 comments
• AI Chatbots Are Like Observational Comics
  by dbreunigon 4/8/2025, 9:07:32 PM with 0 comments
• ChatGPT fixes your DALL-E prompt if it's bad, but lets it be if it's good
  by dbreunigon 3/17/2025, 3:57:07 AM with 0 comments
• Controlling FFmpeg with natural language from the command line
  by dbreunigon 2/24/2025, 8:26:38 PM with 1 comment
• A 16TB Mirror of Data.gov on Source.Coop
  by dbreunigon 2/7/2025, 4:29:22 PM with 7 comments
• A Gentle Guide to Running a Local LLM, for Complete Beginners
  by dbreunigon 2/6/2025, 6:49:57 PM with 1 comment
• DeepSeek Is a Rorschach Test Revealing Each's AI Hopes and Fears
  by dbreunigon 1/31/2025, 6:47:39 PM with 0 comments
• The Rise of the Term "Authentic Data" Illustrates AI Anxiety
  by dbreunigon 1/23/2025, 12:02:15 AM with 1 comment
• Your Eval Is More Important Than the Model: Building an Example Eval
  by dbreunigon 1/9/2025, 6:56:58 PM with 0 comments
• The "Cyren" Song: Meta's AI profiles, Character.ai, and other fake humans
  by dbreunigon 1/3/2025, 4:07:10 PM with 1 comment
• A Primer on Synthetic Data: How It's Helping and Shaping LLMs
  by dbreunigon 12/19/2024, 10:27:21 PM with 0 comments
• DSPy Does the Prompting for You
  by dbreunigon 12/12/2024, 11:20:06 PM with 0 comments
• LLMs Hit the Wall: History Reveals How We Went Fast and Why We'll Go Slow
  by dbreunigon 12/5/2024, 10:58:32 PM with 0 comments
• Detecting DuckDB Requests in Your Webapp to Route Them to Remote DBs
  by dbreunigon 12/1/2024, 9:14:04 PM with 0 comments
• Surging crypto prices dampened interest in AI in 2020. Will they do it again?
  by dbreunigon 11/29/2024, 4:43:44 PM with 0 comments
• Transitioning from the attention era to the automation era
  by dbreunigon 11/18/2024, 9:27:51 PM with 2 comments
• It's the UX, Not the LLM: ChatGPT and Perplexity Can Ship a UI Google Can't
  by dbreunigon 10/31/2024, 8:12:29 PM with 0 comments
• Generating Descriptive Weather Reports with LLMs
  by dbreunigon 10/29/2024, 11:27:08 PM with 0 comments
• AI Use Cases: Gods, Interns, & Cogs
  by dbreunigon 10/18/2024, 8:50:16 PM with 1 comment
• AI PCs Aren't Good at AI: The CPU Beats the NPU
  by dbreunigon 10/16/2024, 7:44:02 PM with 38 comments