When I first launched my Connections Eval (Aug 1st), Gemini 2.5 Pro was able to solve them with no sweat. By August 7th, I had ran quite a few models on this could compare the results including Gemini 2.5 Flash which wasn t so good at 9% solve rate (1 out of 11 puzzles). Continue reading Gemini 3 Flash ruined my eval! The post Gemini 3 Flash ruined my eval! appeared first on DataDuel.co .
I ve been spending time doing some work with AI and this blog is just a check-in journal on what has been working for me. This blog will begin with Idea Honing (mini PRD) and then move to an iterative development workflow, followed by an example project. Idea Honing You need to spend some time Continue reading My exploratory AI workflow The post My exploratory AI workflow appeared first on…
The article discusses the evolution of business intelligence (BI) tools, questioning their longevity compared to the enduring nature of spreadsheets. It highlights how spreadsheets facilitated decision-making in the past but have become inadequate as data complexity increased. The author envisions a future "Spreadsheets 2.0," integrating advanced features, better orchestration, and AI support to…
AI tools like Codium's Windsurf IDE are reshaping coding practices through projects like updating a theme for Evidence.dev and developing a Terraform provider for MotherDuck. These experiences highlighted the strengths and limitations of AI in development, showcasing its potential as a collaborative partner despite issues with project understanding and syntax handling. The post Exploring…
Executive Summary Firmographic data, like industry classification, employee count, and location, is notoriously messy and challenging to analyze. Modern databases offer a unique opportunity to clean and maintain this data: by leveraging an LLM, you can synthesize industry information for your accounts directly within a SQL query. I recently took on exactly this challenge, so Continue reading LLMs…
No one knows what it means, but it s provocative. It gets the people going! Chazz Michael Michaels Context: This is the first of a series of posts looking at Analytics Engineering as I ve lived and breathed it since 2014, largely in the SaaS industry. This is not meant to provide universal truths, but rather give Continue reading Defining Analytics Engineering The post Defining Analytics…
Over the weekend I spent some time getting uv running on mdsinabox.com to see what the hubbub was about. As it turns out, it was harder than expected because of permission issues inside of dev containers github actions. The existing documentation on the uv github repo as well as docker instructions from ryxcommar s blog Continue reading Running uv in dev containers github actions The post Running…
Analysis Narrative Written by Jacob Matson Matt Levine, February 2024. [pdf] Executive Summary Our Methodology Findings Results Raw Frequencies Sample Size: n = 8,162 games Losing Digit Winning Digit Frequency 7 0 3.99% 0 3 3.97% 4 7 3.47% 0 7 3.32% 0 4 3.11% Top 5 most frequent winning squares Weighted Continue reading Analysis of Super Bowl Squares The post Analysis of Super Bowl Squares…
This post was co-written by Alex Noonan, Dani Mermelstein Jacob Matson. Introduction In light of the recent price increases observed in products like dbt Cloud, coupled with the general pressure to maintain cost efficiency, we wanted to build a project template for running dbt in a production environment using GitHub Actions AWS S3. Continue reading Simple dbt runner The post Simple dbt runner…
quick note: the justification for doing this is worth like a 17 page manifesto. I m focusing on the how, and maybe I ll eventually write the manifesto. General Approach This specific problem is loading Point-of-Sale data for a vertical specific system into a database for analysis on a daily basis, but could be generalized to most Continue reading The e in Hex is for ELT The post The e in Hex is…