🎯 Jan 21st Biweekly Clarity AI Meetup
Welcome to our comprehensive guide on AI-powered coding tools, local LLM management, and the latest frameworks for modern development. This page contains key insights from our meetup discussions.
💻 Coding Assistants & IDEs
Cursor
PopularA powerful AI-driven IDE that is a fork of VS Code. Cursor integrates AI deeply into the workflow for building features and managing tickets with advanced context awareness.
- • AI-powered code generation
- • Natural language commands
- • Context-aware suggestions
- • Integrated terminal & git
- • Full-stack development
- • Feature building
- • Rapid prototyping
- • Ticket management
Claude Code
NewA command-line tool and agentic coding assistant from Anthropic that can write code, run tests, and manage git workflows autonomously.
# Install Claude Code
npm install -g @anthropic-ai/claude-code
# Start coding with Claude
claude-code init my-projectContinue
Open SourceAn open-source library/plugin for VS Code and JetBrains that allows users to connect to local LLMs (like Ollama) or remote APIs for AI-assisted coding.
- • VS Code extension
- • JetBrains plugin
- • Ollama support
- • Custom API endpoints
- • Code completion
- • Refactoring assistance
- • Bug fixing
- • Documentation generation
Gemini Code Assist
EnterpriseGoogle's enterprise-grade AI assistance tool integrated into VS Code and other IDEs, designed for professional development teams.
Cloud Code
EvolvingGoogle's suite of IDE extensions evolving quickly to support cloud-based development with Kubernetes, Cloud Run, and more.
- • Kubernetes support
- • Cloud Run integration
- • GCP services access
- • Debug in cloud
Code Companion
NeovimA Neovim plugin that allows users to interact with AI models directly within the terminal-based editor, perfect for vim enthusiasts.
-- Install with lazy.nvim
{
"olimorris/codecompanion.nvim",
dependencies = {
"nvim-lua/plenary.nvim",
"nvim-treesitter/nvim-treesitter",
},
}🔌 Essential VS Code Plugins for AI Development
Continue (AI Code Assistant)
Open-source AI code assistant that works with local LLMs and cloud APIs.
ext install continue.continue
GitHub Copilot
AI pair programmer that suggests code and entire functions in real-time.
ext install GitHub.copilot
Tabnine
AI code completion tool with privacy-focused local model options.
ext install TabNine.tabnine-vscode
Codeium
Free AI-powered code acceleration toolkit with autocomplete and chat.
ext install Codeium.codeium
Gemini Code Assist
Google's enterprise AI coding assistant with advanced features.
ext install GoogleCloudTools.cloudcode
CodeGPT
Connect to multiple AI models including GPT-4, Claude, and local LLMs.
ext install DanielSanMedium.dscodegpt
Ollama
Run local LLMs directly in VS Code with Ollama integration.
ext install ollama.ollama-vscode
AI Commit
Generate meaningful commit messages automatically using AI.
ext install MattIPv4.vscode-ai-commit
⚡ Quick Install Multiple Extensions
code --install-extension continue.continue \
--install-extension GitHub.copilot \
--install-extension TabNine.tabnine-vscode \
--install-extension Codeium.codeium \
--install-extension ollama.ollama-vscode🏠 Running AI Locally - Complete Guide
Ollama
FreeA tool for running large language models (like Llama 3 or DeepSeek) locally on your own machine. Perfect for privacy-conscious development and offline work.
📥 Installation
brew install ollamacurl -fsSL https://ollama.com/install.sh | shDownload from: ollama.com/download/windows
🚀 Quick Start
ollama serveollama pull llama3ollama run llama3🌟 Popular Models
ollama pull llama3
Meta's latest, 8B params
ollama pull deepseek-coder
Specialized for coding
ollama pull mistral
Fast & efficient, 7B params
ollama pull codellama
Code-focused Llama variant
🔗 Connect to VS Code (via Continue)
Configure Continue extension to use Ollama:
{
"models": [{
"title": "Llama 3",
"provider": "ollama",
"model": "llama3"
}],
"tabAutocompleteModel": {
"title": "DeepSeek Coder",
"provider": "ollama",
"model": "deepseek-coder"
}
}LM Studio
GUIA desktop application (primarily Mac, but supports others) that provides a GUI for discovering and running local LLMs without needing the command line.
✨ Key Features
No command line needed
Browse & download models easily
Adjust parameters with sliders
OpenAI-compatible endpoint
📥 Getting Started
- 1. Download from lmstudio.ai
- 2. Install and launch the application
- 3. Browse models in the "Discover" tab
- 4. Download your preferred model (Llama, Mistral, etc.)
- 5. Start chatting or run as a local server
⚖️ Ollama vs LM Studio
| Feature | Ollama | LM Studio |
|---|---|---|
| Interface | Command Line | GUI |
| Platform | Mac, Linux, Windows | Mac, Windows |
| Ease of Use | Requires CLI knowledge | Very beginner-friendly |
| Automation | Excellent for scripts | Limited |
| Best For | Developers, automation | Beginners, experimentation |
🏗️ Frameworks & Infrastructure
Chroma DB
An open-source AI expression database (vector database) used for building RAG (Retrieval-Augmented Generation) systems locally.
- • Document Q&A systems
- • Semantic search
- • Knowledge bases
- • Context-aware AI apps
pip install chromadb
import chromadb
client = chromadb.Client()Linear
A project management tool used to pull in tickets so that AI agents can analyze and build features based on them.
- • API access for agents
- • Ticket analysis
- • Automated feature building
- • Progress tracking
NATS
A connective technology for adaptive edge and cloud-native applications, mentioned for its latest SDK documentation.
Gas Town
BetaA very early-stage/beta project by Steve Yegge focused on Agent Orchestration—having agents manage and verify other agents.
Multi-agent systems where AI agents coordinate, verify each other's work, and manage complex workflows autonomously.
Note: Still in early development stage. Follow Steve Yegge's updates for more information.
Aurora
Mentioned in the context of backend infrastructure and Neovim integration for modern development workflows.
Used for backend infrastructure with Neovim integration, enabling terminal-based development with modern AI capabilities.
📚 Recommended Reading: Vibe Coding
Vibe Coding
By Steve Yegge (with contributions from Gene Kim)
A revolutionary approach to software development where AI handles the syntax and implementation details, while humans focus on the "vibe" or architectural intent of the code. This paradigm shift represents a new way of thinking about coding in the AI era.
🎯 Key Concepts
- • High-Level Thinking: Focus on architecture and design rather than syntax
- • AI as Co-Pilot: Let AI handle implementation details and boilerplate
- • Intent-Driven Development: Communicate what you want, not how to build it
- • Rapid Iteration: Prototype and refine ideas at unprecedented speed
💡 What You'll Learn
- ✓ How to communicate effectively with AI coding assistants
- ✓ Strategies for maintaining code quality with AI assistance
- ✓ When to trust AI vs when to review carefully
- ✓ Building complex systems with AI collaboration
"In vibe coding, you're the architect and the AI is your construction crew. You set the vision, the AI makes it real."
— Steve Yegge
🚀 Quick Start Guide
🌱 Beginner Path (GUI-Focused)
Download from lmstudio.ai and install
Try Llama 3 or Mistral from the Discover tab
Get the Continue extension for VS Code
Use Ctrl+I to ask AI questions in VS Code
⚡ Advanced Path (CLI-Focused)
brew install ollama # or curl method for Linuxollama pull llama3
ollama pull deepseek-coderInstall both and configure to use Ollama
pip install chromadb
# Build your first knowledge base💎 Pro Tips
- 💡 Start with GUI tools (LM Studio) to understand the concepts, then move to CLI (Ollama) for automation
- 💡 Use Continue for VS Code integration - it works with both local and cloud models
- 💡 Keep smaller models (7B-13B) for fast iterations, use larger ones for complex tasks
- 💡 Read "Vibe Coding" to learn the philosophy behind AI-assisted development