Avthar
@avthars
Followers
5K
Following
62K
Media
838
Statuses
10K
Building with AI coding tools and sharing what I learn | prev @timescaledb @Princeton | 🇺🇸🇿🇦
New York
Joined August 2010
NEW AI CODING TUTORIAL -- 6 Months of @claude_code Lessons in 27 Minutes 00:00 @claudeai: The New King of AI Coding? 02:14 Level 1: Beginner 02:29 Installation 03:37 Todo lists 04:16 Essential Commands 06:35 Debugging with Claude Code 07:57 Claude dot md file 09:33 Message Queue
10
106
682
NEW AI CODING TUTORIAL -- 6 Months of @claude_code Lessons in 27 Minutes 00:00 @claudeai: The New King of AI Coding? 02:14 Level 1: Beginner 02:29 Installation 03:37 Todo lists 04:16 Essential Commands 06:35 Debugging with Claude Code 07:57 Claude dot md file 09:33 Message Queue
10
106
682
If you are new to Claude code, here’s the video you need to watch to get up to speed
NEW AI CODING TUTORIAL -- 6 Months of @claude_code Lessons in 27 Minutes 00:00 @claudeai: The New King of AI Coding? 02:14 Level 1: Beginner 02:29 Installation 03:37 Todo lists 04:16 Essential Commands 06:35 Debugging with Claude Code 07:57 Claude dot md file 09:33 Message Queue
1
1
3
Absolute banger from an experienced engineer. Highly suggest you follow Avthar.
NEW AI CODING TUTORIAL -- 6 Months of @claude_code Lessons in 27 Minutes 00:00 @claudeai: The New King of AI Coding? 02:14 Level 1: Beginner 02:29 Installation 03:37 Todo lists 04:16 Essential Commands 06:35 Debugging with Claude Code 07:57 Claude dot md file 09:33 Message Queue
1
1
19
Also on Youtube: https://t.co/zJqsPcIu3J If you want to learn about the best AI coding tools and how to get the most out of them, follow @avthars on X or subscribe on Youtube.
0
0
1
prompting LLMs to do things i could never dream of doing and being annoyed when they can't do it
3
2
50
🫸 STOP COMPACTING, use /clear the biggest difference I see between how the Claude Code team uses CC and the broader community how often they run into compacts instead of keeping long running chats open across multiple tasks, when you finish a task use /clear & start again
60
30
466
Claude Code's GitHub integration is now generally available with a simplified API, ready-to-deploy templates, and support for more GitHub events beyond @-claude mentions.
63
249
3K
Seeing more and more vibe coding platforms targeting mobile apps. This one seems exciting given the founders are Ex-Apple
Today we’re excited to launch Bitrig. Bitrig builds native Swift apps just by chatting with AI. In minutes, and with no code required, you can create an app and deploy it directly to TestFlight. https://t.co/dKkzxZEjFZ
#Swift #SwiftUI #BuildInPublic
0
0
1
How (and why) agents in Claude Code help you write better code, faster: Agents allow you to deploy specialized experts for every task. It's like managing a team rather than collaborating 1-on-1.
80
292
4K
So, I really liked this tweet! It kind of inspired me to just build something for myself. One thing I want is good software for my work as a freelancer to track my expenses and taxes and so on. So, here I am, building my own app for that!
@iannuttall I experienced this earlier this year. The way I overcame it was by building stuff for fun / stuff that I wanted just because it was cool. As @paulg says, “Make something you want”. Don’t be pressured to build something “important” straight away. Once you give yourself this
3
1
2
What it feels like to select o3 in the Legacy Models menu.
52
132
4K
Claude Code as your Devops agent. The new background jobs (like cursors shared terminals) make this way easier. 1. start claude in tmux session 2. ask it to create background process to tail your server logs and give you a summary 3. then ask it to run a second process that
25
77
1K
One piece of news that flew under the radar this week: ChatGPT Pro users can now count Codex CLI toward their usage limits. This mirrors Claude Code's move to allow Claude Max users to use the terminal based coding tool without paying API prices.
3
0
0
Great tips for using Cursor like a pro
here are 31 prompting tips i've collected from the @cursor_ai community 1. write prompts that describe the diff you want in your PR. 2. think of the end state, then prompt for steps to get there. 3. adjust your prompt scope based on results: aim higher if the agent performs
0
0
0
Big things so far from OpenAI GPT-5 livestream: - ChatGPT Free users get GPT-5 (!!) and then gpt-5-mini after hitting limits - Pro users get gpt-5-pro, reasoning model - Best model for coding - Lowest hallucination and error rates of any model (according to new benchmarks)
10
5
25
What can't GPT-5 do / how can it be better? Expanding the dimensions: Computer use, devops, long running (running for days -> weeks -> months on its own)
1
0
0
*Interrupting your GPT-5 news* Cursor releases Claude Code competitor
Cursor is now in your terminal! It’s an early beta. Access all models. Move easily between your CLI and editor.
0
0
0