AI/ML

An AI assistant that remembers and sets goals

Using an LLM to call tools in a loop is the simplest form of an agent. This architecture, however, can yield agents that are “shallow” and fail to plan and act over longer, more complex tasks. Applications like “Deep Research”, “Manus”, and “Claude Code” have gotten around this limitation by implementing a combination of four things: a planning tool, sub agents, access to a file system, and a detailed prompt. Acknowledgements: this exploration was primarily inspired by Claude Code and reports o

A paper from Anthropic describing persona vectors and their applications to monitoring and controlling model behavior

It is wild to think that it has been only a handful of weeks. Claude Code has considerably changed my relationship to writing and maintaining code at scale. I still write code at the same level of quality, but I feel like I have a new freedom of expression which is hard to fully articulate. Claude Code has decoupled myself from writing every line of code, I still consider myself fully responsible for everything I ship to Puzzmo, but the ability to instantly create a whole scene instead of going line by line, word by word is incredibly powerful.

The old timers who built the early web are coding with AI like it's 1995. Think about it: They gave blockchain the sniff test and walked away. Ignored crypto (and …

Everyone's debating whether AI is a bubble while missing the real story. Two things are true: there's a massive AGI fantasy bubble built on geopolitical panic, AND a genuine ML revolution happening at ground level.

The glamourous AI coding agent for your favourite terminal 💘 - charmbracelet/crush

We’re on a journey to advance and democratize artificial intelligence through open source and open science.

Qwen just released their sixth model(!) for this July called Qwen3-Coder-30B-A3B-Instruct—listed as Qwen3-Coder-Flash in their chat.qwen.ai interface. It’s 30.5B total parameters with 3.3B active at any one time. This means …

Some of my attempts to make agents work better that just didn’t work.

The Moonshot AI team synthesized thousands of tools, agents, users, and sessions to build a library of training data.

Contribute to T3-Content/SnitchBench development by creating an account on GitHub.

Aeneas, a generative neural network trained on ancient texts, helps historians contextualize inscriptions and perform epigraphic tasks, offering an improved starting point for historical research.

ai.bythebay.io Nov 2025, Oakland, full-stack AI conference DSPy: Prompt Optimization for LM Programs Michael Ryan, Stanford It has never been easier to build amazing LLM powered applications. Unfortunately engineering reliable and trustworthy LLMs remains challenging. Instead, practitioners should build LM Programs comprised of several composable calls to LLMs which can be rigorously tested, audited, and optimized like other software systems. In this talk I will introduce the idea of LM Programs in DSPy: The library for Programming — not Prompting LMs. I will demonstrate how the LM Program abstraction allows the creation of automatic optimizers for LM Programs which can optimize both the prompts and weights in an LM Program. I will conclude with an introduction to MIPROv2: our latest and highest performing prompt optimization algorithm for LM Programs. Michael Ryan is a masters student at Stanford University working on optimization for Language Model Programs in DSPy and Personalizing Language Models. His work has been recognized with a Best Social Impact award at ACL 2024, and an honorable mention for outstanding paper at ACL 2023. Michael co-lead the creation of the MIPRO & MIPROv2 optimizers, DSPy’s most performant optimizers for Language Model Programs. His prior work has showcased unintended cultural and global biases expressed in popular LLMs. He is currently a research intern at Snowflake.

This app explains how language models transform text into meaningful representations through embeddings. It provides a visual guide to help you understand traditional and modern language model tech...

A practical guide to working effectively with AI coding tools like Claude Code, covering mindset shifts, quality control strategies, and team collaboration workflows for modern software development.

Your robot experience started simple. You typed a question into a chatbot… just to see. Can it answer that question? I'd be impressed if it did. Your query was simple. A simple knowledge question that with a little effort using legacy tools like Google, you would have discovered yourself, but the r

Claude Code Subagents Collection. Contribute to davepoon/claude-code-subagents-collection development by creating an account on GitHub.

Practical lessons from building production agentic systems

VS Code AI Customization - Learn to use custom instructions, prompt files, and custom chat modes to personalize AI code generation, reviews, and chat responses.

claude code system prompt. Contribute to kn1026/cc development by creating an account on GitHub.

Diffusion models, CLIP, and the math of turning text into images Welch Labs Book: https://www.welchlabs.com/resources/imaginary-numbers-book Sections 0:00 - Intro 3:37 - CLIP 6:25 - Shared Embedding Space 8:16 - Diffusion Models & DDPM 11:44 - Learning Vector Fields 22:00 - DDIM 25:25 Dall E 2 26:37 - Conditioning 30:02 - Guidance 33:39 - Negative Prompts 34:27 - Outro 35:32 - About guest videos + Grant’s Reaction Special Thanks to: Jonathan Ho - Jonathan is the Author of the DDPM paper and the Classifier Free Guidance Paper. https://arxiv.org/pdf/2006.11239 https://arxiv.org/pdf/2207.12598 Preetum Nakkiran - Preetum has an excellent introductory diffusion tutorial: https://arxiv.org/pdf/2406.08929 Chenyang Yuan - Many of the animations in this video were implemented using manim and Chenyang’s smalldiffusion library: https://github.com/yuanchenyang/smalldiffusion Cheyang also has a terrific tutorial and MIT course on diffusion models https://www.chenyang.co/diffusion.html https://www.practical-diffusion.org/ Other References All of Sander Dieleman’s diffusion blog posts are fantastic: https://sander.ai/ CLIP Paper: https://arxiv.org/pdf/2103.00020 DDIM Paper: https://arxiv.org/pdf/2010.02502 Score-Based Generative Modeling: https://arxiv.org/pdf/2011.13456 Wan2.1: https://github.com/Wan-Video/Wan2.1 Stable Diffusion: https://huggingface.co/stabilityai/stable-diffusion-2 Midjourney: https://www.midjourney.com/ Veo: https://deepmind.google/models/veo/ DallE 2 paper: https://cdn.openai.com/papers/dall-e-2.pdf Code for this video: https://github.com/stephencwelch/manim_videos/tree/master/_2025/sora Written by: Stephen Welch, with very helpful feedback from Grant Sanderson Produced by: Stephen Welch, Sam Baskin, and Pranav Gundu Technical Notes The noise videos in the opening have been passed through a VAE (actually, diffusion process happens in a compressed “latent” space), which acts very much like a video compressor - this is why the noise videos don’t look like pure salt and pepper. 6:15 CLIP: Although directly minimizing cosine similarity would push our vectors 180 degrees apart on a single batch, overall in practice, we need CLIP to maximize the uniformity of concepts over the hypersphere it's operating on. For this reason, we animated these vectors as orthogonal-ish. See: https://proceedings.mlr.press/v119/wang20k/wang20k.pdf Per Chenyang Yuan: at 10:15, the blurry image that results when removing random noise in DDPM is probably due to a mismatch in noise levels when calling the denoiser. When the denoiser is called on x_{t-1} during DDPM sampling, it is expected to have a certain noise level (let's call it sigma_{t-1}). If you generate x_{t-1} from x_t without adding noise, then the noise present in x_{t-1} is always smaller than sigma_{t-1}. This causes the denoiser to remove too much noise, thus pointing towards the mean of the dataset. The text conditioning input to stable diffusion is not the 512-dim text embedding vector, but the output of the layer before that, [with dimension 77x512](https://stackoverflow.com/a/79243065) For the vectors at 31:40 - Some implementations use f(x, t, cat) + alpha(f(x, t, cat) - f(x, t)), and some that do f(x, t) + alpha(f(x, t, cat) - f(x, t)), where an alpha value of 1 corresponds to no guidance. I chose the second format here to keep things simpler. At 30:30, the unconditional t=1 vector field looks a bit different from what it did at the 17:15 mark. This is the result of different models trained for different parts of the video, and likely a result of different random initializations. Premium Beat Music ID: EEDYZ3FP44YX8OWT

GitHub Spark was released in public preview yesterday. It’s GitHub’s implementation of the prompt-to-app pattern also seen in products like Claude Artifacts, Lovable, Vercel v0, Val Town Townie and Fly.io’s …

Natural Language Web. Contribute to microsoft/NLWeb development by creating an account on GitHub.

We built a browser-based terminal controller in one day using Claude Code, named pipes, and Xterm.js. No SSH needed, just open your browser and start typing. Check and command your agents on the go!

I’m a little salty that neither Anthropic nor Google reached out to me before they released their terminal-based AI coding agents.

Will McGugan is building his own take on a terminal coding assistant, in the style of Claude Code and Gemini CLI, using his Textual Python library as the display layer. …