Kristian Muñiz

@krismuniz.com

149 Followers 492 Following

Software developer, designer, and open source enthusiast. 🇵🇷 Technologist turned software engineer by necessity. Building software products https://krismuniz.com

A wide image taken with a phone of a glass whiteboard, in a room overlooking the Bay Bridge. The field of view shows a woman writing, sporting a tshirt wiith a large OpenAI logo. The handwriting looks natural and a bit messy, and we see the photographer's reflection. The text reads: (left) "Transfer between Modalities: Suppose we directly model p(text, pixels, sound) [equation] with one big autoregressive transformer. Pros: * image generation augmented with vast world knowledge * next-level text rendering * native in-context learning * unified post-training stack Cons: * varying bit-rate across modalities * compute not adaptive" (Right) "Fixes: * model compressed representations * compose autoregressive prior with a powerful decoder" On the bottom right of the board, she draws a diagram: "tokens -> [transformer] -> [diffusion] -> pixels"
The video showcases a web-based, interactive Python coding environment. Three editor nodes in an infinite canvas user interface display: 1) Python code using matplotlib to generate a bar chart (initially "Serious Chart"), 2) the chart's visual output with export options, and 3) Python code calculating cosine similarity with its numerical output. The video demonstrates real-time updates. The user modifies the first Python script, changing the chart title to "Yearly Clown Shoe Sales" and data, causing the chart on the second canvas to update instantly. The user then adds text output to the cosine similarity script, which also updates in real-time on the third canvas. The interface dynamically renders both graphical and textual Python outputs directly within the browser.
This video demonstrates coding in TypeScript within an infinite, dark-themed canvas environment. Three code windows showcase different TypeScript features: one defines a "Person" interface and logs an instance, demonstrating type checking and code completion; another illustrates DOM interaction and browser API access using document.getElementById and window.navigator.canShare along with type hints; and the last one highlights npm support by importing and using the "validator" (for email validation) and "redash" (for string transformation) libraries. The user interacts with the code by typing, deleting, and utilizing autocompletion suggestions, highlighting the dynamic nature of the interface, and drags the code windows across the scrollable canvas. A control panel in the top right corner offers customization options.