If you're still reading code the old way, I have news for you I've onboarded onto codebases with zero documentation more times than I'd like to admit. I grepped around, read the README (if it exists), and looked at past commits. That old approach is outdated. Here are 4 AI tools that help you understand codebases: 1️⃣ DeepWiki: Replace GitHub. com with deepwiki. com in any repo URL. You get auto-generated wiki docs, architecture diagrams, and an AI chat that actually knows the code. Built by the team behind Devin: deepwiki [dot] com 2️⃣ Code Wiki: Gemini-powered living documentation that regenerates after every commit. Every section hyperlinks to the actual code. The chat knows your entire repo end-to-end. Built by Google: codewiki [dot] google 3️⃣ GitSummarize: Turn any GitHub repo into a full documentation hub with summaries and high-level overviews. Built because the creators found it difficult to understand massive codebases when trying to contribute to open source. Free with rate limits: gitsummarize [dot] com 4️⃣ Code2Tutorial: Paste a GitHub URL, get a step-by-step tutorial walkthrough. Think of it as an AI-generated guided course for any repo: code2tutorial [dot] com Alternatively, just ask your favorite coding agent (e.g. Codex / Claude Code), it does a decent job of navigating your codebase. #frontenddeveloper #reactjsdeveloper #webdeveloper #developer #frontend #reactjs #nextjs #redux #ai -copied

If you use Claude Code, you've probably noticed longer implementations burn through your context fast. So I built a plugin that hands the implementation off to OpenCode. Claude reads the repo, writes a plan, you say yes, OpenCode builds it with whatever model you want — glm, kimi, gpt-5, local ollama, whatever you have. Claude verifies. Give it a try and let me know what you think. A star would mean a lot. https://lnkd.in/dGxBkjwN

I’m incredibly excited to announce the open-source release of LLMFront! 🚀 I was getting frustrated with having to juggle different interfaces—one for testing Hugging Face models in the cloud, and entirely different clunky setups for running offline AI securely. So I built my own self-hosted, unified solution from scratch using Python (FastAPI) and Vanilla JavaScript. With LLMFront, anyone can explore open-source AI without blowing up their PC's memory. Some of the most fun backend/frontend engineering challenges I solved: 🧠 Semantic Search with Llama3-Groq: You can now query the UI with natural language (e.g., "I want a coding AI") and it instantly matches and filters the exact models from the Hugging Face hub. 🖥️ Out-Of-Memory (OOM) Protection: Built an integrated VRAM profiler that checks hardware capacities and dynamically warns you or forces 4-bit/8-bit quantization before loading massive models. 🌍 Dynamic Bilingual Isolation: A custom real-time i18n engine that translates the entire UI and System Prompts seamlessly without heavy framework refreshes. ☁️ Native Ollama endpoint emulation and resilient background downloads with heavy-cache protections. The codebase is entirely free and designed with local hardware in mind. I would absolutely love to hear what the community thinks and welcome any Pull Requests! 🔗 GitHub Repository: https://lnkd.in/dVk2DrZV #ArtificialIntelligence #OpenSource #FastAPI #HuggingFace #Python #WebDev #Groq

46% of developers picked Claude Code as their most loved AI dev tool. Cursor got 19%. Copilot got 9%. I switched from Cursor six months ago. Didn't plan to. I was debugging a race condition and got tired of copy-pasting context between the terminal and the IDE like a human clipboard. So I stayed in the terminal. Haven't gone back. The thing I was wrong about: I thought the IDE was my development environment. It wasn't. It was a middleman. IDE plugins only see what the extension API exposes. Terminal tools inherit your entire machine. Git history, running containers, MCP servers, CI pipelines. I now run 4-6 Claude Code sessions simultaneously using git worktrees. Each with its own branch, its own scope. One refactors the API layer while another writes tests. Try doing that in an IDE without losing your mind. IDEs aren't dying though. I still open VS Code for visual debugging and pair programming. But the creative work, the architecture decisions, the "figure out what's wrong and fix it" work? That's terminal now. The IDE got demoted from operating system to viewer. And I don't think it's going back. Full breakdown with benchmarks and the workflow patterns that changed how I build: https://lnkd.in/db24NB8u #ClaudeCode #DeveloperTools #AgenticAI

Ever stared at a massive GitHub repository and wished you could just ask it how it works? I’ve been diving deep into local AI orchestration and wanted a better way to navigate new codebases. So, I built GitChat, A Retrieval-Augmented Generation (RAG) assistant that lets you chat directly with any public GitHub repository. It dynamically clones a repo, chunks the architecture, and streams context-aware answers back to a sleek, minimalist dark-mode interface. The best part? The LLM inference runs entirely locally. Features: - Dynamic Repo Ingestion: Automatically clones, filters out heavy/binary files, and chunks code for AI ingestion. - Zero-Hallucination Guardrails: Strict prompt engineering ensures the AI only answers using the provided codebase context. - Local LLM Orchestration: Runs models (like Qwen 2.5 Coder) locally, meaning zero API costs for inference. - Real-Time Streaming UI: Seamless, non-blocking message streaming with syntax-highlighted markdown rendering. - Dynamic Sandbox Controls: Real-time adjustments for LLM Temperature, K-value context retrieval, and model selection. The Tech Stack: - Frontend & API: Next.js (App Router), Tailwind CSS - AI Engine: Vercel AI SDK v6, LangChain - Database & Vector Search: MongoDB Atlas Vector Search - Local Models: Ollama (qwen2.5-coder, nomic-embed-text) I'm really proud of the resource management and the seamless integration between the backend data pipeline and the frontend UI. GitChat is fully open-source, and I am actively welcoming contributions! I would love for you to check out the architecture, test it out, or even open a PR if you have ideas for new features. Repository & Code: https://lnkd.in/dNuBfnzZ #Nextjs #LangChain #Ollama #MongoDB #SoftwareEngineering #AI #WebDevelopment

Day 21 of #30AppIn30Days — I built APIBlueprint 🔷 Describe your API in plain English → get a full OpenAPI 3.0 spec in seconds. I've designed a lot of APIs. The actual design is the interesting part. Writing the YAML isn't. So I built a tool that handles the boilerplate: → Type what your API does in plain English → Llama 3 (via Groq's free API) generates a complete OpenAPI 3.0 spec → Visual explorer shows all endpoints grouped by tag — expandable with params, request body, and response codes → Schema browser for all your data models → Syntax-highlighted raw view with line numbers → Export to openapi.json or openapi.yaml — drop straight into Postman or Swagger UI The whole app runs in the browser. No backend. No account. Just React + Vite + a free Groq API key. The part I'm most proud of technically: I wrote the syntax highlighter and YAML serializer from scratch (~50 lines total) instead of pulling in libraries. Sometimes the lightweight solution is the right one. Stack: React PWA · Vite · Llama 3.3 70B · Groq API · JetBrains Mono 🔗 Code in the comments 21 apps down. 9 to go. #buildinpublic #react #openapi #devtools #llm #groq #indiedev #30AppIn30Days

I built a 15,400-line full-stack application without writing most of the code myself. Not a todo app. StockPulse — an intelligent stock analysis platform. Two scoring engines, 40+ React components, 16 database tables, 30+ API endpoints, 11 test suites. Built entirely with Claude Code. But the real leverage wasn't the tool — it was the system I built around it: MCP Integrations — Context7 pulling official docs for Next.js 16, Bun, and TailwindCSS v4 directly into context. Playwright running browser-based happy path tests. No guessing. No stale knowledge. Best Practices as Skills — Framework-specific best practices verified against official documentation, codified into reusable skills. Claude doesn't "hope" it knows the right pattern — it loads verified, current guidance on demand. No hallucinated patterns. No deprecated approaches. Custom Skills — /scaffold-route, /add-endpoint, /implement-flow, /code-review, /test-coverage-gaps. Each one a markdown file version-controlled through Git. Team conventions encoded, not remembered. File 90 follows the same standards as file 1. Hooks — PostToolUse hooks running Prettier and ESLint after every edit. PreToolUse hooks blocking destructive operations. Stop hooks auto-running tests after every response. SessionStart hooks injecting project context into every session. Quality enforcement that never sleeps, never forgets, never gets tired at midnight. At line 15,000, the code follows the same patterns as line 100. Because best practices were codified as skills, hooks enforced quality automatically, and MCP integrations grounded every decision in verified docs. Claude Code isn't a coding assistant. It's a force multiplier — if you invest in the system around it. Without that, it's expensive autocomplete. #ClaudeCode #AI #BuildInPublic #NextJS #TypeScript #DeveloperProductivity

I built InsightOps Review Hub, a full-stack internal knowledge operations application for turning private source material into reviewable, citation-backed briefing answers. The project demonstrates a complete controlled AI workflow: - Workspace-based ownership and access control - Source document storage and deterministic excerpt generation - Citation-backed answer generation without external API keys - Normalized citation records instead of loose JSON references - Persisted automated evaluation runs - Separate human review decisions - Audit trail visibility across core workflow actions - Typed React + TypeScript frontend - Django REST Framework backend - SQLite local setup and PostgreSQL via Docker Compose - Backend tests, frontend tests, Playwright smoke flow, and GitHub Actions CI The goal was not just to build CRUD, but to model how AI-assisted internal tools can stay traceable, reviewable, and operationally credible. Tech stack: Django, Django REST Framework, React, TypeScript, Vite, PostgreSQL, SQLite, Docker, GitHub Actions, Vitest, Playwright. GitHub: https://lnkd.in/edURbXtY #SoftwareEngineering #AIEngineering #Django #React #TypeScript #FullStackDevelopment #Docker #PostgreSQL #InternalTools #HumanInTheLoop #AIGovernance

Introducing Spandrel: an open source context engineering pipeline that turns scattered information into a knowledge graph served via MCP and GraphQL. Just tell your coding agent:  “Clone github(.)com/trevorfox/spandrel.git then read BOOTSTRAP.md and follow its instructions to set up my knowledge graph.” The agent reads your existing content, proposes a structure, and builds a knowledge graph you can query in minutes. You confirm what looks right, adjust what doesn't, and walk away with a compiled graph that any agent can navigate via MCP. Why I built it:  Every team building with AI agents has the same problem. The knowledge exists — in docs, Slack, notes, people's heads — but there's no structured way for agents to access it. They burn tokens loading everything or miss critical context searching for fragments. One command to set up. Markdown files in a git repo. A compiler that builds the graph. GraphQL as the universal query layer. MCP tools and skills that any agent can use. No vector database. No embeddings pipeline. No new paradigm. Just the technologies you already use, turned into powerful agent capabilities. Open source, MIT licensed: https://lnkd.in/gURKVvdW

Your GitHub is already your portfolio. You’re just not using it that way. Most developers: • manually pick projects • rewrite descriptions • maintain separate portfolios And it gets outdated… fast. So I built something to fix that. 👉 GitProfolio — https://lnkd.in/gZ3v6hKJ It converts your GitHub into a structured, portfolio-ready JSON. Not just raw data — it actually understands your work: • Ranks your projects based on impact (stars, forks, recency) • Extracts clean descriptions from messy READMEs • Detects your real tech stack automatically • Finds live project links (Vercel, Netlify, etc.) • Highlights your top projects instantly And if you want more: It can also generate insights to: • improve your repo visibility • suggest better descriptions & topics • give project ideas that attract stars No more manually maintaining portfolios. Your GitHub becomes your portfolio — automatically. Built using Astro, GitHub API, and AI-powered analysis. Would love to hear what you think 👇 #GitHub #WebDevelopment #DeveloperTools #OpenSource #Portfolio #BuildInPublic #IndieHackers #JavaScript #AI #DevTools