All Posts

A practical implementation-level walkthrough of how Ragas uses schema-constrained LLM judging (structured outputs + retries) and how key RAG metrics compute their final scores.

A satirical look at why skipping AI agent evaluations makes perfect sense if you don't value maintainability, customers, or long-term sanity.

How to model query quality, use bootstrapping, and report realistic RAG performance with RAGAS metrics and confidence intervals.

MLE-Bench introduces a new benchmark to evaluate AI agents on real-world ML engineering tasks using Kaggle competitions. This post highlights key findings, including resource scaling effects, debugging challenges, and the performance of different agent frameworks.

This post explores a neural network model designed for the Causal Discovery Challenge organized by ADIA Lab. It highlights the use of a Transformer-based architecture with two layers of scaled dot-product attention and layer normalization, achieving a multi-balanced accuracy of 47.986%.