Removing one section of a small ontology at a time, then letting the model write its own. The joins did most of the work, and the model could write those, but not the store's own meanings.
Sixteen questions built on words with a store-specific meaning, asked with and without the definitions. Made-up words were declined. Familiar words were often answered with the model's own meaning.
A small test with a model that runs on a laptop: readable schemas, hidden names, and what happens when the model is allowed to say it can't answer.
Key AI ethics concepts from Georgia Tech OMSCS AIES (CS 6603): data and privacy, statistical literacy, fairness metrics, and bias mitigation.
Key knowledge-based AI concepts from Georgia Tech OMSCS KBAI (CS 7637): knowledge representation, reasoning strategies, learning, planning, and the ARC-AGI project.
A few months into team-wide AI use, here are the patterns I'm seeing, and the question that keeps coming back about what should stay consistent across the team.
A walkthrough of Claude Code's execution lifecycle, from the moment you type a prompt to the final response
Moving from a Copilot-only environment to a team that actively uses Claude Code: what changed and what I'm still figuring out
How I introduced a WiX-based Windows installer for deploying field software to air-gapped environments, and the unexpected challenges along the way.
Key software engineering concepts from Georgia Tech OMSCS SDP (CS 6300): lifecycle models, UML, testing, Android development, and team collaboration.
Key robotics AI concepts from Georgia Tech OMSCS AI4R (CS 7638): localization, planning, perception, control, and SLAM.
Key ML-for-trading concepts from Georgia Tech OMSCS ML4T (CS 7646): portfolio theory, signals, risk, and evaluation.
Explore core concepts of unsupervised learning, including K-means clustering, optimization strategies, and how anomaly detection systems are designed and evaluated.
A complete guide to decision trees, covering entropy, information gain, one-hot encoding, regression trees, and ensemble methods like Random Forest and XGBoost.
Learn how to make decisions, evaluate models, handle bias and variance, and manage real-world ML workflows with cross-validation, error analysis, and transfer learning.
Explore how neural networks are trained with gradient descent, softmax, and backpropagation using TensorFlow. Understand activation functions and multiclass classification techniques.
A comprehensive guide to neural networks, forward propagation, TensorFlow implementation, and efficient matrix computations.
A comprehensive breakdown of logistic regression, sigmoid function, loss functions, and regularization for classification tasks.
Deep dive into multiple linear regression, vectorization, gradient descent, feature scaling, and polynomial regression.
Overview of supervised and unsupervised learning, linear regression, and gradient descent