Agentic AI, LLM engineering, and the models behind modern automation — multi-agent systems, LLM evaluation and comparisons, RAG, fine-tuning, AI infrastructure, security, and production AI engineering.
From the blog
Practical examples of automating Microsoft Office applications with UFO — from Excel data manipulation and Word document formatting to Outlook email workflows, with multi-step task descriptions and result verification.
Learn how to build an end-to-end document intelligence agent that combines Tesseract OCR, layout detection, zone classification, and structured information extraction to process any document type automatically.
Build an AI agent that analyzes architectural floor plans to detect rooms, classify their types, estimate areas, identify furniture, and generate natural language descriptions for real estate and interior design applications.
Learn how to build an AI agent for medical image analysis that preprocesses X-rays and scans, detects findings, generates structured reports, and includes appropriate clinical disclaimers for responsible deployment.
Master cross-application automation with Microsoft UFO by building workflows that transfer data between Excel, Outlook, browsers, and desktop applications through coordinated multi-app sequences.
Build a production-grade price monitoring agent that scrapes multiple e-commerce sites, extracts prices with AI, detects changes, sends alerts, and maintains a historical price database for trend analysis.
Design agent teams that improve collectively through experience sharing, collective memory, skill transfer, and performance benchmarking. Includes Python implementations for experience replay and cross-agent learning.
Learn to build an AI agent for social media automation covering platform API integration versus browser automation, content scheduling, engagement monitoring, and rate limiting strategies.
Build a visual question answering agent that understands images, routes questions to specialized analysis modules, performs multi-modal reasoning, and generates accurate natural language answers about any image content.
