Members-Only
Recent Talks & Demos are for members only
You must be an AI Tinkerers active member to view these talks and demos.
Playwright MCP: Deep Agent RPA
Learn to build robust, parallelizable RPA using LLMs, Pydantic Deep-Agents, and Playwright MCP for efficient web automation.
How do we build RPA that doesn’t take more time fixing it, than writing it? By using LLMs! Let me show you how I built out sessionized, distinct, parallelisable RPA process using pydantic deep-agents, the Playwright MCP, and a little bit of creativity to take web automation to the next level!
- pydantic-deepDeep merge Pydantic models and dictionaries recursively while maintaining strict schema validation.Standard Pydantic models lack native deep merge capabilities for nested structures. This library fills that gap by recursively updating nested BaseModels and dictionaries. It ensures partial updates to complex schemas remain type-safe and validated. Use it to handle API PATCH requests or layered configuration files where shallow updates would overwrite entire objects. It supports Python 3.8+ and integrates directly with Pydantic v1 and v2 workflows.
- PlaywrightPlaywright is the Microsoft-developed, cross-browser automation framework: it drives Chromium, Firefox, and WebKit with one unified API for fast, reliable end-to-end testing.Playwright delivers robust, cross-platform end-to-end testing, supporting all major rendering engines: Chromium, Firefox, and WebKit. Launched by Microsoft in January 2020, its core strength is a single API for multiple languages (TypeScript, Python, Java, .NET). The framework eliminates flaky tests through automatic waiting and provides full test isolation by creating a new browser context (a brand-new browser profile) for each test. Key tooling includes Codegen for recording actions and the Trace Viewer for deep post-mortem analysis of test failures (screencasts, live DOM snapshots). This architecture ensures reliable, high-speed execution across Windows, Linux, and macOS.
- MCP serversMCP Servers (Model Context Protocol) standardize bi-directional communication, connecting Generative AI agents to external enterprise data and tools with secure, low-latency interfaces.MCP Servers are the critical bridge for agentic AI: they implement the Model Context Protocol, an open standard developed by Anthropic, to expose specific capabilities to AI applications. These servers manage data communication, ensuring LLMs access fresh context from source systems like databases or file systems, maintaining conversational latency for real-time responses. They operate using a client-server model, utilizing JSON Schema for tool validation, enabling agents (e.g., Claude, GitHub Copilot) to execute complex, multi-step tasks like querying a PostgreSQL database or managing a GitHub pull request, all through a single, standardized interface.
- ChromiumChromium is the open-source web browser project providing the codebase for Google Chrome, Microsoft Edge, and numerous other major browsers.Chromium is the foundational, open-source browser project, primarily developed by Google, that drives a significant portion of the modern web. This C++ codebase provides the vast majority of source for Google Chrome, plus major competitors like Microsoft Edge, Opera, and Brave. The project utilizes a multi-process architecture, a critical sandbox security model, the Blink rendering engine, and the V8 JavaScript engine for high-speed performance. Licensed under the BSD-3 clause, Chromium operates on a rapid, fixed release schedule, pushing a new stable version approximately every six weeks: its development model is a proven, high-velocity platform for secure, stable, and standards-compliant web experiences across Windows, macOS, Linux, and Android.
- DockerDocker is the open-source platform that packages applications and dependencies into standardized, portable containers for consistent execution across any environment.Docker is the industry-standard containerization platform, enabling developers to build, ship, and run applications efficiently. It uses the Docker Engine (the core runtime) to create lightweight, isolated environments called containers: these units bundle an application’s code, libraries, and configuration. This self-contained approach guarantees consistency, eliminating the 'it works on my machine' problem across development, testing, and production environments (local workstations, cloud, or on-premises). Docker debuted in 2013 and now serves over 20 million developers monthly, simplifying complex workflows like CI/CD and microservices architecture by leveraging tools like Docker Hub for image sharing and Docker Compose for multi-container applications.
Related projects
Playwright MCP vs CLI? Benchmarking for performance & effectiveness
Denver
Discover surprising performance trade-offs between Playwright MCP and CLI. This talk reveals how interface design impacts agent efficiency…
How AI Agents Actually Work With Tools and MCP
Boston
A walkthrough of the actual code flow behind AI agents, demonstrating how they invoke tools like the Multi‑Tool…
Coding with AI: What Works & What Doesn't
Manchester Nh
Learn practical AI coding assistant dos and don'ts, avoiding slowdowns. See a live demo of a self-learning agent…
Dual-Agent MCP Development: Builder and Tester in the Same Conversation
Seattle
See a live demo of building a 44-tool MCP server using two collaborating Claude agents: one builds, one…
MCP – Future of agents
Poland
The talk explains the MCP protocol, its operation, and demonstrates practical use cases, illustrating how it enables advanced…
Let AI Agents Dogfood Their Code With WebMCP
Seattle
This talk introduces WebMCP, a fork of the Chrome DevTools MCP server, enabling AI agents to call frontend…