AI / ML — Page 7 of 9
Browsing page 7 of 9 — 99 articles on ai / ml.
Building a Document Processing Pipeline with AI: OCR, Classification, and Extraction for Production Systems
How to build a production document processing system that ingests PDFs, images, and scanned documents, classifies them by type, extracts structured data, and routes downstream. Covers OCR engine selection, embedding-based classification, LLM extraction with schemas, confidence scoring, and human-in-the-loop review queues.
Building a Voice AI Pipeline: Speech-to-Text, LLM Processing, and Text-to-Speech in Production
A production-focused deep dive into end-to-end voice AI architecture covering audio capture, STT provider selection, LLM streaming, TTS synthesis, WebSocket real-time delivery, latency optimization, and cost modeling with TypeScript examples throughout.
Building an AI-Powered Code Review Pipeline: Static Analysis, LLM Review, and Automated Feedback
How to build an automated code review pipeline that combines static analysis with LLM-powered review: webhook ingestion, diff parsing, staged analysis, comment posting, prompt design, token budget management, and calibrating when to suppress automated feedback.
AI-Powered Testing: LLM-Generated Test Cases, Visual Regression, and Autonomous QA Pipelines
How to use LLMs to generate meaningful test cases, build visual regression systems that understand intent, and construct autonomous QA agents that navigate real applications. Includes TypeScript examples, architecture patterns, and the real production limitations you need to plan around.
Building a Recommendation Engine: Collaborative Filtering, Vector Similarity, and Real-Time Personalization in Production
Recommendation systems in production are rarely just matrix factorization. This guide covers collaborative filtering, embedding-based similarity, cold start strategies, real-time vs batch pipelines, feature stores, A/B testing, and low-latency serving architecture with TypeScript examples.
Building LLM-Powered Internal Tools: Natural Language to SQL, Document Q&A, and Workflow Automation
A practical guide to building LLM-powered internal tools that work in production. Covers natural language to SQL, document Q&A with RAG and access control, and LLM-driven workflow automation, with TypeScript examples and honest failure mode analysis.
AI Agent Evaluation in Production: Testing Frameworks, Quality Gates, and Regression Detection
How to build a rigorous evaluation harness for AI agents: offline test sets, online monitoring, CI/CD quality gates, regression detection, and A/B testing agents in production.
Knowledge Graphs for RAG: When Vector Search Alone Is Not Enough
Vector similarity retrieval breaks on multi-hop reasoning, entity relationships, and temporal ordering. This guide covers where vector-only RAG fails, how to build and query a knowledge graph, and how to combine both in a hybrid retrieval system with TypeScript examples and production tradeoffs.
LLM Function Calling in Production: Tool Routing, Input Validation, and Error Recovery Patterns
Function calling is how LLMs take actions in the world. Done naively, it creates silent failures, security holes, and unpredictable costs. This article covers the full production picture: schema design, input validation before execution, parallel vs sequential calls, error recovery, security, observability, and when to skip function calling entirely.
Agentic Engineering in Practice: Governing AI-Generated Code at Scale
What agentic engineering actually means beyond the hype, and the governance layer humans must own when AI agents generate most of your production code.
Building an Internal AI Assistant: RAG Over Private Documents with Access Control and Audit Trails
How to build a production internal AI assistant that answers questions over private company documents while enforcing per-user access control, maintaining compliance audit trails, and handling document lifecycle without leaking information across permission boundaries.
Deploying Open-Source LLMs in Production: Model Serving, Quantization, and Infrastructure Choices
A practical guide to running Llama, Mistral, and Qwen in production. Covers model serving frameworks (vLLM, TGI, Ollama), quantization tradeoffs (GPTQ, AWQ, GGUF), GPU provisioning, batching strategies, load balancing, and when self-hosting beats managed API providers.