Ai

Sun-ha Hong's Tech Policy Press essay argues that tech hype consumes real resources and narrows society's choices, leaving behind failed infrastructure while recycled promises, inflated forecasts, and concentrated power crowd out alternatives.

Elastic Ramen brings Elastic Agent Builder into the terminal for SRE incident investigation and remediation.

llmfit is a CLI/web tool that inspects your machine's CPU, RAM, and GPU/VRAM, then scores which local LLMs will actually run well on that hardware by memory fit, speed, quality, and context length.

A Citrini Research report modeling an 'AI crisis' scenario in which agentic AI displaces high-earning white-collar workers, breaking the feedback loop between productivity gains and consumer spending. It projects 10.2% U.S. unemployment and a 38% cumulative S&P 500 drawdown by mid-2028.

A 2023 Medium article comparing the text detection and recognition performance of four open-source OCR engines: Tesseract, EasyOCR, PaddleOCR, and MMOCR.

A Hugging Face Space by user Loren for comparing OCR output from EasyOCR, PaddleOCR, MMOCR, and Tesseract on an uploaded image.

A Medium article by Steve Yegge introducing Beads, a git-backed, graph-based issue-tracker built to give coding agents persistent memory across sessions. It was released in October 2025 to address AI coding agents losing context between work sessions.

A January 2026 Steve Yegge Medium post introducing Gas Town, an orchestration system for running many Claude Code agent instances in parallel.

Anthropic's product page for Claude Code, an agentic coding tool that runs in the terminal or an IDE. It can read and edit a codebase, run commands, and carry out multi-step coding tasks.

TheFocus.AI is a Brooklyn-based AI consulting studio that works alongside client teams under a Strategy/Build/Teach model — finding where AI fits, building it into real workflows, and teaching the team to keep it running. Its product, Habitats, is a shared workspace where a team's AI work is visible and reusable.

An article examining the Books1 and Books2 datasets used to train large language models, noting they may overlap substantially with the BookCorpus dataset of roughly 11,000 books. It discusses how Books1 is believed to draw from public-domain works while Books2's contents remain undocumented.

Lisanne Bainbridge's 1983 paper 'Ironies of Automation', hosted with commentary, argues that automating a process does not remove its difficulties but shifts them onto the human operators left to handle the remaining exceptions.

Aequitas is an open-source bias and fairness audit toolkit for machine learning models, developed by the Center for Data Science and Public Policy at the University of Chicago. Its docs cover installing the Python package, using its CLI, and reading the fairness metrics it generates across subgroups to support equitable decision-making.

Alibi is an open-source Python library for machine learning model inspection and interpretation, maintained by SeldonIO. It implements local and global explanation methods such as Anchor, Counterfactual, Integrated Gradients, and ALE for tabular, text, and image data.

Deepchecks is an open-source platform for validating machine learning data and models, from tabular data to NLP and computer vision. It groups individual validation checks into suites covering data integrity, distribution drift, and model performance.

DeepMind Lab is a 3D research environment built on the Quake III Arena engine, used as a testbed for reinforcement learning agents on navigation and puzzle-solving tasks.

Fairlearn is an open-source Python toolkit for assessing and improving fairness in machine learning systems. It pairs an interactive dashboard with unfairness-mitigation algorithms for navigating trade-offs between fairness and model performance.

Great Expectations is an open-source Python framework for validating, profiling, and documenting data quality. Assertions about a dataset, called Expectations, are expressed as human-readable Python methods and run against data sources ranging from CSV files to databases.

MLflow is an open-source platform for the machine learning and AI application lifecycle, covering experiment tracking, model versioning, evaluation, and deployment. It is backed by the Linux Foundation and integrates with over 100 tools across the ML ecosystem.

TensorFlow's guide to TensorFlow Model Analysis (TFMA), a library for evaluating TensorFlow models at scale within a TFX pipeline. It computes metrics over data slices and can visualize the results in a Jupyter notebook.

WhyLabs, an AI observability platform for monitoring machine learning models and LLM applications, discontinued operations in 2026. Its full platform was open-sourced, and its whylogs and LangKit tools continue on as open-source projects.

Yellowbrick is a Python library of diagnostic visualizations that extends scikit-learn's API to help steer feature engineering, algorithm selection, and hyperparameter tuning during model selection.

Kaggle's dataset hub, where individuals, research groups, and organizations publish and share open datasets for machine learning and data analysis. Listings typically include descriptions, data dictionaries, and discussion threads.