Deep Learning, NLP, NMT, AI, ML
At a glance MagenticLite is an agentic application that works across both the browser and local file system in a single workflow. Built as the next generation of Magentic-UI, it combines a redesigned app with a harness optimized for small models. MagenticBrain and Fara1.5 are small models designed for orchestration and computer-use tasks, respectively. Fara1.5 is the next iteration of Fara and
Read moreAt a glance Vega lets users prove facts from government-issued credentials — age, personhood, professional status — without revealing the credential itself. The credential never leaves the device. Zero-knowledge proofs are generated in under 100 ms on
Read moreInteractive Quiz ⋅ 9 QuestionsBy Joseph Peart Share In this quiz, you’ll test your knowledge of How to Use the Claude API in Python. By working through this quiz, you’ll revisit how to install the anthropic SDK, send prompts to Claude with client.messages.create(), shape responses with a system parameter, and return structured JSON output using a schema or Pydantic. The quiz contains 9 questions and there is no time limit. You’ll get 1 point for each correct answer. At the […]
Read moreThe fastest way to use the Claude API in Python is to install anthropic, set your API key, and call client.messages.create(). You’ll have a working response in under a minute: Example of Using the Claude API in Python Claude is Anthropic’s large language model, accessible via a clean REST API with an official Python SDK. Unlike heavier AI frameworks that require you to wire up multiple components before
Read moreInteractive Quiz ⋅ 8 QuestionsBy Joseph Peart Share In this quiz, you’ll test your understanding of Tapping Into the Zen of Python. By working through this quiz, you’ll revisit the origins of the poem, the meaning of several aphorisms, and the inside jokes hidden throughout. The questions explore how the principles apply in practice and when it’s okay to bend the rules in the name of practicality. The quiz contains 8 questions and there is no time limit. You’ll get […]
Read moreInteractive Quiz ⋅ 7 QuestionsBy Joseph Peart Share In this quiz, you’ll test your understanding of Absolute vs Relative Imports in Python. By working through this quiz, you’ll revisit how Python’s import system resolves modules, the differences between absolute and relative imports, and the PEP 8 conventions for styling import statements. The quiz contains 7 questions and there is no time limit. You’ll get 1 point for each correct answer. At the end of the quiz, you’ll receive a total […]
Read moreThe Zen of Python is a collection of 19 aphorisms that capture the guiding principles behind Python’s design. You can display them anytime by running import this in a Python REPL. Tim Peters wrote them in 1999 as a joke, but they became an iconic part of Python culture that was even formalized as PEP 20. By the end of this video course, you’ll understand: The Zen of Python is a humorous poem of 19 aphorisms describing Python’s design philosophy […]
Read morePython’s built-in functions are predefined functions you can use anywhere in your code without any imports. They handle common tasks across math, data type creation, iterable processing, and input and output. Knowing which ones to reach for makes your code shorter and more Pythonic. To get the most out of this tutorial, you’ll need to be familiar with Python programming, including topics like working with built-in data types, functions, classes, decorators, scopes, and the import system. Built-in Functions in Python […]
Read more
For a long time, “scaling” in foundation models mostly meant one thing: spend more compute on pre-training and capabilities rise. That intuition was supported by empirical work such as Kaplan et al. (2020), which reported predictable power-law trends in loss as you scale model parameters, dataset size, and training compute. In practice, these trends justified sustained investment in large-scale accelerator capacity and the surrounding distributed infrastructure needed to keep it efficiently utilized. But the frontier has evolved—and scaling is no […]
Read more
TL;DR: we explain how to separate CPU and GPU workloads to get a massive performance boost for inference. This is the second post in a series on efficient LLM inference. The first post covered continuous batching from first principles. It introduces some concepts we build upon: KV cache, FlashAttention, attention masks, etc. An H200 costs around $5 an hour on Inference Endpoints. That’s cheap for an hour, but use it for a day and you are already paying $120. If […]
Read more