The editorial argues that three curriculum repos (freeCodeCamp, free-programming-books, project-based-learning) dominating GitHub trending contradicts the dominant VC/CEO narrative that junior developer roles are being eaten by AI agents. Star velocity on these repos is a leading indicator that people are still entering the field in growing numbers, despite claims the economics no longer make sense.
The project-based-learning repo's curated list of 'build a thing to learn a language' tutorials embodies the position that learners retain skills by shipping real projects, not consuming abstract theory. Its 272k stars and continued star velocity signal broad developer endorsement of learning-by-building as the entry model.
freeCodeCamp's open-source curriculum has offered a project-heavy, structured path into software for over a decade. Its position at 437.9k stars — the top of GitHub trending — reinforces the argument that a free, curriculum-driven, hands-on route into the field remains highly valued even in the agent era.
The free-programming-books repo takes the position that a volunteer-maintained markdown list of freely available programming books is a legitimate and enduring foundation for developer education. Its 384k stars and ongoing star velocity indicate that people entering the field still overwhelmingly value curated, free, community-vetted reading lists over paid bootcamps or paywalled content.
The everything-claude-code repo positions itself as an agent harness performance optimization system for Claude Code, Codex, Opencode, and Cursor. Its 115.1k stars represent the counter-position that the trending page's real story is the explosion of agent-adjacent tooling, not curriculum repos.
The opencode repo is an open source coding agent trending at 118.5k stars, arguing implicitly that agent-based development tooling is where developer attention is actually flowing. Its presence high on trending supports the AI-is-eating-junior-work narrative that the editorial pushes back against.
AutoGPT continues to trend at 182.3k stars with its vision of accessible AI for everyone to build on. Its sustained momentum represents the position that autonomous agent frameworks — not learning resources — are what developers are actively bookmarking and building around.
LangChain markets itself as 'the agent engineering platform' at 128.9k stars, taking the position that agent construction is now a first-class engineering discipline. Its presence on trending argues that the era's real skill demand is building with agents, not learning to code from scratch.
GitHub's trending page this week is doing something quietly strange. The top three repos by star velocity aren't AI agents, aren't new frameworks, aren't LLM wrappers. They're freeCodeCamp (437.9k stars), EbookFoundation/free-programming-books (384k stars), and practical-tutorials/project-based-learning (272k stars). Three curriculum repos, sitting on top of a trending list that's supposed to be a leading indicator of where developer attention is going.
None of them shipped a headline feature this week. freeCodeCamp is the same open-source curriculum it's been for a decade. `free-programming-books` is a markdown list of PDFs and links, maintained by volunteers. `project-based-learning` is exactly what its name suggests: a curated list of "build a thing to learn a language" tutorials. The kind of repo that would have been unremarkable in 2018 and is, in 2026, apparently the most-starred content on the platform this week.
The interesting question isn't why these repos exist — it's why their star velocity is spiking now, in the same month that every VC deck insists junior developer roles are being eaten by Claude Code and Devin.
The dominant narrative for the last eighteen months has been that entry-level coding work is collapsing. Google's CEO said 25% of new code is AI-generated. Amazon rolled out an internal agent that closes tickets end-to-end. Marginalia's Viktor Lofgren argued last week that the engineers most at risk aren't juniors but the mid-career seniors who stopped learning after their bootcamp. Every one of those takes assumes the funnel of people *entering* the field is shrinking, or should be, because the economics no longer make sense.
The trending page disagrees.
Star velocity on GitHub is a noisy signal, but it's a signal. It measures what developers — and, increasingly, people who want to become developers — are bookmarking, forking, and sharing this week. When three curriculum repos aggregating over a million stars between them all trend simultaneously, that's not one viral tweet. That's sustained interest from a population that hasn't gotten the memo about the death of the junior role.
A more honest read: the pool of people learning to code from scratch is not shrinking. It may be growing. The barrier to entry — free curriculum, free books, project-based tutorials in every language — has never been lower. And the incentive to *start* has, if anything, gone up: even if agents write half the code, someone has to specify the system, review the diffs, and know when the agent is lying. You cannot review code you never learned to read.
There's also a second-order effect worth naming. `free-programming-books` and `project-based-learning` are both essentially *lists of links*. Their value proposition is human curation. In a world where ChatGPT will happily hallucinate a book that doesn't exist and a tutorial URL that 404s, a hand-maintained list of things that provably work becomes more valuable, not less. The AI content flood is a tailwind for hand-curated GitHub lists, not a headwind.
freeCodeCamp is a slightly different story. It's not just a list — it's a full interactive curriculum with certifications, a job board, and a nonprofit behind it. Its persistence at the top of the trending page for years suggests something the industry rarely admits out loud: the free-tier learning path still works. People go through it, get jobs, come back, contribute. That flywheel doesn't care about your token pricing.
If you're hiring, the practical takeaway is that the top-of-funnel for junior candidates is healthier than the LinkedIn discourse suggests. The people showing up to your open req didn't stop existing — the industry stopped opening the reqs. That's a hiring-side decision, not a supply-side collapse. When budgets loosen, the pipeline is still there.
If you're mentoring or writing docs, the trending signal tells you where attention lives: project-based, cite-able, hand-curated. The docs that get shared aren't the ones that read like reference manuals. They're the ones structured as "build X to learn Y." `project-based-learning` didn't hit 272k stars by being comprehensive. It hit it by being scannable. If your internal onboarding doc reads like an API reference, it's the wrong format.
If you're building tools for developers, you're probably still building for the median mid-career engineer. The trending page is telling you there's a much larger population one rung below that, and they're picking their tools right now — starting with the free ones. Getting into a `free-programming-books` recommendation is worth more long-tail traffic than a paid ad. That's not a growth hack; that's just how the funnel actually works.
The agent hype cycle isn't wrong about capability — Claude Code and its peers really do ship code. But it's confused about the labor market. The number of people who want to learn to write code is not a function of whether AI can also write code; it's a function of whether coding remains a path to a decent living, and the answer is still yes. The trending page is a weekly reminder that the pipeline into this profession is wider than the discourse thinks, and the people entering it are still, in 2026, reaching for freeCodeCamp and a curated markdown file first. The question worth watching isn't whether they'll finish the curriculum. It's what job market they'll graduate into eighteen months from now.
freeCodeCamp.org's open-source codebase and curriculum. Learn math, programming, and computer science for free.
→ read on GitHub:books: Freely available programming books
→ read on GitHubYour own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞
→ read on GitHubA list of Free Software network services and web applications which can be hosted on your own servers
→ read on GitHubSkills for Real Engineers. Straight from my .claude directory.
→ read on GitHubThe agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
→ read on GitHubCurated list of project-based tutorials
→ read on GitHubThe agent that grows with you
→ read on GitHubLinux kernel source tree
→ read on GitHubThe open source coding agent.
→ read on GitHubFair-code workflow automation platform with native AI capabilities. Combine visual building with custom code, self-host or cloud, 400+ integrations.
→ read on GitHubA feature-rich command-line audio/video downloader
→ read on GitHub🔥 The API to search, scrape, and interact with the web for AI
→ read on GitHubAn Open Source Machine Learning Framework for Everyone
→ read on GitHubOpen-source Windows and Office activator featuring HWID, Ohook, TSforge, and Online KMS activation methods, along with advanced troubleshooting.
→ read on GitHubVisual Studio Code
→ read on GitHubA curated list of awesome Go frameworks, libraries and software
→ read on GitHubAutoGPT is the vision of accessible AI for everyone, to use and to build on. Our mission is to provide the tools, so that you can focus on what matters.
→ read on GitHubGet up and running with Kimi-K2.5, GLM-5, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.
→ read on GitHubFlutter makes it easy and fast to build beautiful apps for mobile and beyond
→ read on GitHubf.k.a. Awesome ChatGPT Prompts. Share, discover, and collect prompts from the community. Free and open source — self-host for your organization with complete privacy.
→ read on GitHubThe most popular HTML, CSS, and JavaScript framework for developing responsive, mobile first projects on the web.
→ read on GitHub🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.
→ read on GitHubProduction-ready platform for agentic workflow development.
→ read on GitHubA modern GUI client based on Tauri, designed to run in Windows, macOS and Linux for tailored proxy experience
→ read on GitHubLangflow is a powerful tool for building and deploying AI-powered agents and workflows.
→ read on GitHubThe agent engineering platform
→ read on GitHubThe most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface.
→ read on GitHubCollection of publicly available IPTV channels from all over the world
→ read on GitHubThe React Framework
→ read on GitHubMicrosoft PowerToys is a collection of utilities that supercharge productivity and customization on Windows
→ read on GitHubProduction-Grade Container Scheduling and Management
→ read on GitHub:electron: Build cross-platform desktop apps with JavaScript, HTML, and CSS
→ read on GitHubNode.js JavaScript runtime ✨🐢🚀✨
→ read on GitHubEmpowering everyone to build reliable and efficient software.
→ read on GitHub This project is dedicated to collecting high-quality macOS software and organizing them systematically by different categories for easy search and use.
→ read on GitHubBuild smaller, faster, and more secure desktop and mobile applications with a web frontend.
→ read on GitHubA modern runtime for JavaScript and TypeScript.
→ read on GitHubTop 10 dev stories every morning at 8am UTC. AI-curated. Retro terminal HTML email.