The article titled "A Google Team Measured Half of My Argument, and Left the Other Half Open" is part of a series on spec‑driven test automation (Part 3). It discusses how a Google team evaluated only half of the author’s argument, leaving the remaining portion unresolved. The post was originally published on the Towards Data Science platform.
The article titled "Towards Spec-Driven Test Automation: Part 2" discusses the implications of a single test run and what it actually proves. It appears on the Towards Data Science platform.
The true bottleneck was never the analysis. The post BI Is Dead, Long Live BI appeared first on Towards Data Science .
The article "Towards Spec-Driven Test Automation: Part 1" discusses how a green test suite may not truly reflect software quality. It explores the limitations of relying solely on test pass rates and introduces the concept of specification-driven testing as a more robust approach. The post is published on Towards Data Science.
A tiny GPT model was trained to memorize every fact presented to it, yet it failed to answer any question that required combining two of those facts. The author tested a one-line fix suggested in a recent paper, which only partially resolved the issue. The post highlights the limitations of small language models in integrating multiple pieces of information.
Checking an A/B test until it crosses p < 0. 05 can turn a nominal 5 percent false-positive rate into almost 28 percent.
Enterprise Document Intelligence [Vol. 1 #8C] - Structured output is the start of validation, not the end: check the evidence, accept not-found, loop the feedback The post Validating the RAG Answer Before the User Sees It: Spans, Quotes, and the Feedback Loop appeared first on Towards Data Science .
The barriers to building have collapsed. That shifts the bottleneck to ownership, validation, taste, and deciding what should actually exist The post Code Is Cheap.
The article discusses the challenge of measuring the impact of AI systems, noting that obtaining data is now easier than deciding how to interpret it. It highlights the difficulty of turning raw numbers into actionable insights. The piece was originally published on Towards Data Science.
The article discusses a statistical paradox illustrating how combining data can reverse the narrative presented by individual charts. It explains that both charts can be correct yet appear contradictory when aggregated. The piece highlights the importance of careful data integration to avoid misleading conclusions.
Enterprise Document Intelligence [Vol. 1 #9B] - One call wires the four upgraded bricks together, run on a paper, a NIST standard, and a report with a broken TOC The post A Production RAG Pipeline in Action: Every Answer Typed and Cited appeared first on Towards Data Science .
The article recounts a production incident where a large language model (LLM) was used to evaluate the outputs of another LLM, and the judging model consistently agreed with itself. It explores the implications of relying on one model to assess another’s work, highlighting the potential pitfalls of such an approach. The narrative offers lessons on the limits of trusting automated evaluation systems in real‑world deployments.
I tried to make my ETL pipeline production-ready. Three things broke.