Members-Only
Recent Talks & Demos are for members only
You must be an AI Tinkerers active member to view these talks and demos.
Fabraix: Technical Browser Agent Security
Explore Fabraix's architecture, showing how a Chrome extension isolates a local AI agent, securely captures page context, and enforces real-time security checks.
This demo is a technical show-and-tell of how we built Fabraix, a browser-native AI assistant designed from the ground up to be trustworthy. I’ll walk through the architecture of our Chrome extension, focusing on the interplay between the local agent running in the browser and our core security engine. We’ll dive into the code that captures page context in real-time, how we process it securely, and the design decisions we made to prevent the agent from being hijacked, even when connected to sensitive web apps. Expect a live demo of the agent completing a complex task, with a walkthrough of the specific security checks happening at each step.
Fabraix secures enterprise AI agents from unpredictable behavior and and attacks.
- PythonPython: The high-level, general-purpose language built for readability, powering everything from web backends to advanced machine learning models.Python is the high-level, general-purpose language prioritizing clear, readable syntax (via significant indentation), ensuring rapid development for any team . Its ecosystem is massive: use it for robust web development with frameworks like Django and Flask, or leverage its power in data science with libraries such as Pandas and NumPy . The Python Package Index (PyPI) provides thousands of community-contributed modules, offering immediate solutions for tasks from network programming to GUI creation . The language is actively maintained by the Python Software Foundation (PSF), with the stable release currently at Python 3.14.0 (as of November 2025) .
- TypeScriptTypeScript is an open-source superset of JavaScript: it adds static typing and compiles to clean, standards-based JavaScript.TypeScript is a high-level, open-source language developed by Microsoft: it acts as a superset of JavaScript, adding a powerful static type system. This system enables compile-time type checking, catching errors before runtime (a critical benefit for large-scale applications). The TypeScript Compiler (TSC) reliably transpiles all code into clean, standards-based JavaScript (ES3 or newer), ensuring compatibility across any browser or host environment (Node.js, React.js, etc.).
- FastAPIFastAPI is a modern, high-performance Python web framework for building APIs with automatic OpenAPI documentation.FastAPI is a robust, high-speed Python web framework: it is built on Starlette (for async capabilities) and Pydantic (for data validation and serialization). Leveraging standard Python 3.8+ type hints, the framework automatically generates interactive API documentation (Swagger UI/ReDoc) and enforces data validation, effectively reducing developer-induced errors by an estimated 40%. This architecture delivers performance on par with Node.js and Go, significantly increasing feature development speed (up to 300% faster). It is production-ready, fully supporting OpenAPI and JSON Schema standards for all API specifications.
- GeminiGoogle's natively multimodal AI model: understands and operates across text, code, audio, image, and video.Gemini is Google's most capable and general AI model, engineered from the ground up to be natively multimodal: it seamlessly understands and combines information across text, code, audio, image, and video inputs. The technology is optimized for flexibility, running efficiently on everything from data centers to mobile devices. It is deployed in three key sizes: Ultra (for highly complex tasks), Pro (for broad scaling), and Nano (for efficient on-device tasks). Developers access this power via the Gemini API to build next-generation applications.
- Google CloudGlobal-scale cloud computing: The secure, AI-powered platform running on the same infrastructure as Google Search and YouTube.Google Cloud (GCP) is a comprehensive suite of cloud computing services (IaaS, PaaS, and SaaS), providing the core infrastructure for digital transformation. It operates on Google's global network, spanning over 30 regions and 90 zones, ensuring high availability and low latency . Key products include Compute Engine (VMs), Google Kubernetes Engine (GKE), BigQuery (serverless data warehouse), and Vertex AI (for enterprise-ready generative AI) . GCP enables organizations to build, deploy, and scale applications efficiently, leveraging Google's expertise in security, data analytics, and cutting-edge machine learning . New customers receive $300 in free credits, plus access to over 25 products for free, up to monthly usage limits .
Related projects
Private Agents in your browser
Chicago
Demonstrates how to run small-language-model agents directly in the browser, enabling low-latency, cost-effective agentic tasks on a laptop…
Building 'mission control' for software engineering agents
London
Live demo of Ona's architecture for parallel software engineering agents, showing isolated, ephemeral environments, secure code execution, and…
Our in-house analytics agent & Lessons learned from agent-to-agent communication
London
A walkthrough of building a self‑serve BI tool using Streamlit on EC2 with Claude‑generated queries, followed by real‑world…
Browser-use agent inside a frontend component
San Francisco
See a frontend component agent implementation with custom loops and UI tools. Learn unique context engineering strategies beyond…
Fun Hacks for Chat Interfaces
London
Discover techniques to collapse Vercel resumable streams, embed context in Anthropic API calls while keeping cacheability, and sanitize…
Building Conversational AI Agents
London
Learn practical steps to design, develop, and deploy conversational AI agents, covering architecture, language models, training data, evaluation,…