MachineLearningMastery.com

31 articles in the Truth Foundry index. Each links out to the original.

• Run Small Language Models Locally on Your Machine in 15 Minutes with Ollama · Ollama simplifies local AI by packaging models into a lightweight service for offline inference.
• Scikit-Ollama Enables Local LLM Zero-Shot Classification Without Cloud APIs · Scikit-Ollama bridges scikit-learn syntax with local Ollama models for zero-shot text classification.
• Comparing RAGAS, DeepEval, and Promptfoo for LLM Evaluation · Guide to selecting and implementing open-source frameworks for evaluating LLM applications.
• LangChain vs LlamaIndex vs Raw API Calls: Choosing the Right LLM Abstraction · A technical comparison of LangChain, LlamaIndex, and raw API calls for building LLM applications
• Guide to Choosing Between AI Tools and Subagents Without Overengineering · A framework for deciding when to use tools versus subagents in AI agent architecture.
• Six Techniques to Fix AI Agent Tool Selection Accuracy at Scale · Retrieval-based tool selection triples agent accuracy by filtering tool catalogs before model inference.
• Understanding Context and Memory Engineering in AI Agents · This article explains how context engineering and memory engineering solve distinct problems in agentic AI systems and their interaction at the retrieval boundary.
• Seven Layers of the Production AI Agent Tech Stack Explained · The AI agent stack comprises seven layers from foundation models to deployment infrastructure.
• Agentic Workflows vs. Autonomous Agents: Control Flow Distinctions Explained · Agentic workflows use predefined paths while autonomous agents dynamically control their own execution flow.
• Build Text Clustering Pipeline Using LLM Embeddings and HDBSCAN · Tutorial demonstrates clustering unstructured text using LLM embeddings, UMAP, and HDBSCAN.
• Building Browser-Using AI Agents in Python with Playwright and LangGraph · Tutorial on creating AI agents that browse websites using Playwright, browser-use, and LangGraph.
• Mastering AI Agent Evaluation: Beyond Final Outputs · Rigorous AI agent evaluation requires examining the full execution process, not just final results.
• Effective AI Agent Tool Design Patterns Prevent Common Failures · Reliable AI agents require single-responsibility tools, tight schemas, and structured error handling.
• Five Essential Python Concepts for Scalable AI Engineering Systems · AI engineers must master generators, context managers, async programming, Pydantic, and magic methods for production systems.
• Perform Multi-Label Text Classification with Scikit-LLM and Groq LLMs · Learn to use scikit-LLM for zero-shot multi-label text classification with Groq-hosted LLMs
• Building Multimodal AI in the Browser with Transformers.js · This article teaches how to build browser-based multimodal AI for image and speech tasks using Transformers.js without servers or APIs.
• AgentOps: A Practitioner's Guide to Observability for Autonomous AI Agents · AgentOps provides session replay, cost governance, and safety guardrails for production AI agents.
• Build Client-Side Semantic Search with Transformers.js and Sentence Embeddings · Learn to create a serverless semantic search engine using Transformers.js and cosine similarity.
• Benchmarking TF-IDF, BART, and Scikit-LLM for Text Classification · A tutorial compares classical TF-IDF pipelines, BART zero-shot models, and Scikit-LLM for text classification tasks.
• Six-Step LLMOps Roadmap for Building Production-Grade LLM Systems in 2026 · A structured six-step roadmap covers observability, evaluation, cost control, and agent orchestration for production LLMs.
• Continuous Batching Solves LLM Inference Bottlenecks with Dynamic Scheduling · Dynamic and ragged batching eliminate GPU idle time caused by static request padding in LLM inference.
• Implementing Context Pruning for Long-Running AI Agents Using Semantic Similarity · A tutorial on using sentence transformers to prune LLM context by retaining only semantically relevant past turns.
• Understanding LLM Token Selection: Logits, Temperature, and Top-P Sampling · LLMs use logits, temperature, and top-p sampling to control output creativity and predictability.
• Building a Resilient Multi-Tool Gemma 4 Agent with Error Recovery · This tutorial teaches how to build a resilient Gemma 4 agent that recovers from tool failures and malformed outputs.
• Implementing Hybrid BM25 and Semantic Search in RAG Systems · Tutorial on combining BM25 lexical and semantic search using Reciprocal Rank Fusion for RAG
• Python Tutorial Builds Context-Aware Semantic Search with LLM Embeddings · Tutorial demonstrates building a context-aware search engine in Python using sentence embeddings and metadata filtering.
• Build Multi-Agent AI Research Assistant with Python, OpenAI SDK, and Olostep · Tutorial on creating a Python-based multi-agent research assistant using OpenAI Agents SDK and Olostep web APIs.
• Roadmap to Building Production-Grade Agentic AI Systems · A structured guide to mastering agentic programming from foundational concepts to shipping production agents.
• Prompt Engineering for Agentic AI: Principles and Patterns for Reliable Behavior · Prompt engineering for agentic AI requires context engineering and four core components for reliable autonomous behavior.
• Implement Semantic Search in PostgreSQL Using pgvector Extension · PostgreSQL's pgvector extension enables semantic similarity search using vector embeddings and SQL.