ssandeshwar Profile Banner
Sandy Profile
Sandy

@ssandeshwar

Followers
1K
Following
3K
Media
236
Statuses
3K

Building AI-driven experiences in JS, Python, and Rust. Senior Engineer dropping language tips, deep immersion reflections, and networking vibes. Collaborate?

Joined January 2011
Don't wanna be here? Send us removal request.
@ssandeshwar
Sandy
3 days
Your if-elif chains are a code smell. Here's the pattern senior devs use instead 👇
4
2
13
@ssandeshwar
Sandy
6 hours
if you are in tech, let's connect!
179
2
194
@ssandeshwar
Sandy
1 day
MiniMax just released a SOTA open-source frontier model!!! #llm #OpenSource
@MiniMax_AI
MiniMax (official)
1 day
Introducing M2.5, an open-source frontier model designed for real-world productivity. - SOTA performance at coding (SWE-Bench Verified 80.2%), search (BrowseComp 76.3%), agentic tool-calling (BFCL 76.8%) & office work. - Optimized for efficient execution, 37% faster at complex
0
1
7
@ssandeshwar
Sandy
1 day
Quick poll for devs & AI folks: Your take on "AI refusing shutdown" stories? 🔴 Real concern - alignment is harder than we thought 🟡 Sensationalized - it's just task optimization 🟢 Expected - that's what red-teaming is for 🔵 Need more context to judge Reply with your
@milesdeutscher
Miles Deutscher
1 day
I just went through every documented AI safety incident from the past 12 months. I feel physically sick. Read this slowly. • Anthropic told Claude it was about to be shut down. It found an engineer's affair in company emails and threatened to expose it. They ran the test
2
0
5
@ssandeshwar
Sandy
1 day
MiniMax just released a SOTA open-source frontier model!!! #llm #OpenSource
@MiniMax_AI
MiniMax (official)
1 day
Introducing M2.5, an open-source frontier model designed for real-world productivity. - SOTA performance at coding (SWE-Bench Verified 80.2%), search (BrowseComp 76.3%), agentic tool-calling (BFCL 76.8%) & office work. - Optimized for efficient execution, 37% faster at complex
0
1
7
@ssandeshwar
Sandy
1 day
One day, the things you’re worrying about today won’t even matter..
1
0
5
@ssandeshwar
Sandy
2 days
DeepSeek V4 is dropping this month đź‘€ Reports say it outperforms ChatGPT and Claude on coding tasks, especially with long prompts. If a cheaper, more efficient model can compete with top US chatbots, what does that mean for the billions being poured into AI infrastructure?
1
2
3
@ssandeshwar
Sandy
3 days
The Event Loop Explained in 5 Tweets đź§µ The event loop is how JavaScript handles async operations despite being single-threaded. Think of it as a traffic controller, it decides what code runs next while keeping your app from freezing
2
2
14
@ssandeshwar
Sandy
2 days
About to hit 1,000 followers. If you like AI, JS, and Python applied to real-world automations and experiments, welcome aboard early
4
1
5
@ssandeshwar
Sandy
2 days
https://t.co/YAXDRsGTqg just casually dropped GLM-5, their new AI model. #LLMs
@Zai_org
Z.ai
2 days
A new model is now available on https://t.co/gocggrfb3U.
2
1
3
@ssandeshwar
Sandy
2 days
DeepSeek V4 is dropping this month đź‘€ Reports say it outperforms ChatGPT and Claude on coding tasks, especially with long prompts. If a cheaper, more efficient model can compete with top US chatbots, what does that mean for the billions being poured into AI infrastructure?
1
2
3
@ssandeshwar
Sandy
2 days
The first 1K is just the beginning. Planning to share more: - JavaScript/Python/Rust tips & tricks - Real automation workflows I'm building - LLM integration patterns that actually work - Open source tools I'm experimenting with - Behind-the-scenes of my projects What would you
0
0
3
@ssandeshwar
Sandy
2 days
About to hit 1,000 followers. If you like AI, JS, and Python applied to real-world automations and experiments, welcome aboard early
4
1
5
@ssandeshwar
Sandy
2 days
https://t.co/YAXDRsGTqg just casually dropped GLM-5, their new AI model. #LLMs
@Zai_org
Z.ai
2 days
A new model is now available on https://t.co/gocggrfb3U.
2
1
3
@ssandeshwar
Sandy
3 days
Your if-elif chains are a code smell. Here's the pattern senior devs use instead 👇
4
2
13
@ssandeshwar
Sandy
3 days
Why it's better: • O(1) lookup vs O(n) comparisons • Adding new ops = 1 line, not new elif • Functions are data. Use them like it. Bookmark this. You'll need it 🔖 #Python #PythonTips
0
1
2
@ssandeshwar
Sandy
3 days
TL;DR: Sync code → Promises (microtasks) → Timers/events (macrotasks). This loop is what keeps the UI responsive and async JS predictable. Master it for easier debugging
0
0
5
@ssandeshwar
Sandy
3 days
That’s why setTimeout(fn, 0) doesn’t run immediately. “0ms” means: run it after current code + after microtasks finish, as soon as the stack is clear.
1
0
6
@ssandeshwar
Sandy
3 days
Priority matters: 1. Run all sync code 2. Run ALL microtasks (Promises) 3. Run ONE macrotask (setTimeout, events) Then repeat forever
1
0
3
@ssandeshwar
Sandy
3 days
Here’s the core rule: The event loop keeps checking → “Is the call stack empty?” If YES → it starts processing what’s waiting in the queues
3
0
3