Amit Tripathi

Co-Founder at Shorteefy | We built an AI video clipper that beats a $10M/yr competitor

[Amit Tripathi @ce_amit]

Tired of Generic Clips from AI video clippers?

No more.

Introducing Shorteefy (@iamshorteefy) - an AI video clipper that repurposes long-form content into short-form clips without losing your creative voice.

Whether you want Control or Speed, Shorteefy adapts to how you create.

We have just launched. Give it a try and let us know what you think.

[Amit Tripathi @ce_amit]

Claude Fable के जाने का ग़म थोड़ा कम हुआ सा लग रहा है, GLM 5.2 के बारे में अच्छी बातें सुनकर।

Open source community को सलाम। 🙏

[Amit Tripathi @ce_amit]

अच्छी सैटिएटी (तृप्ति) देते हैं।

ऐसे खाद्य पदार्थ आपको लंबे समय तक भरा हुआ महसूस कराते हैं, धीरे-धीरे ऊर्जा देते हैं और आम तौर पर स्वास्थ्य के लिए भी बेहतर होते हैं।

[Amit Tripathi @ce_amit]

सैटिएशन चेक (Satiation Check)

लंच के 3 घंटे बाद, अपनी सबसे पसंदीदा खाने की चीज़ के बारे में सोचिए।

अब खुद से पूछिए:

"अगर यह चीज़ अभी मेरे सामने रख दी जाए, तो क्या मेरा इसे खाने का मन करेगा?"

अगर जवाब "नहीं" है, तो संभव है कि आपका लंच ऐसे खाद्य पदार्थों से बना था जो .....

[Amit Tripathi @ce_amit]

@MaziyarPanahi Could you please put some light on memory requirements? What if I want to run it with a CPU, what are the minimum requirements?

[Amit Tripathi @ce_amit]

@MaziyarPanahi How is better than OpenAI Privacy Filter?

[Amit Tripathi @ce_amit]

I have stopped using Windsurf when they have changed the pricing structure.

Earlier it was discrete and I could control how much I want to use a model.

Now, daily and weekly limits and I do not know how I am using it

If you could give both options - @windsurf - all is good.

[Swagat Nayak @beingcarrot]

Update on my previous post: I ordered a ₹3 Lakh RTX 5090 on @AmazonIN and got a 1.56kg packet of detergent. Amazon promised a refund to kill the social media buzz, but they're just stalling. Now, I've uncovered a massive internal FBA fraud ring.

Thread @AmazonHelp @AmitAgarwal

[Amit Tripathi @ce_amit]

I have just added a skill for weaknesses in same. It helps a lot to learn the repo and then ways to improve it.

[Zara Zhang @zarazhangrui]

Codebase-to-course now has 2.6k stars on GitHub

Just optimized it to be a lot more token efficient & reliable

Originally intended for vibe coders to learn CS, but was told it's great for developer onboarding as well

[Amit Tripathi @ce_amit]

Leonie has been sharing a lot about Agentic workflows. I have registered.

[Leonie @helloiamleonie]

Hope to see you this Wednesday! 👋

Sign up here: https://t.co/z43KOqOZ6x

[Amit Tripathi @ce_amit]

I recently found a 5 Rs Good Day biscuits packet which is way tastier than any bigger size.

It is small in size, baked deep and dark, goes really well with tea.

[Amit Tripathi @ce_amit]

Triggers Anxiety

[Pankaj @the2ndfloorguy]

i built one for myself, dramatically playing AajTak news music before every meeting 🚨

[Amit Tripathi @ce_amit]

I use Paytm and I thought it is most used, but it is behind Google Pay too. Surprising.

[Indian Tech & Infra @IndianTechGuide]

🚨 The evolution of UPI market share in India.

[Amit Tripathi @ce_amit]

Moved to round 2. Fingers crossed!

[Amit Tripathi @ce_amit]

@IGPcom Thanks IGP. I got the resolution.

[Amit Tripathi @ce_amit]

@IGPcom I have reverted back. Thanks

[Amit Tripathi @ce_amit]

After a while I have started using OpenAI model.

Too good.

[Matt Shumer @mattshumer_]

I've been testing GPT-5.4 for the last week.

In short, it is the best model in the world, by far.

It's so good that it's the first model that makes the “which model should I use?” conversation feel almost over.

The biggest surprise: I barely use Pro anymore!

If you know me, you know I'm a Pro addict. I reach for Pro models constantly, and use them for almost everything, as they just... nail almost anything I give to them.

For the first time, 5.4's standard version, with heavy thinking, just broke that habit.

Even in standard mode, GPT-5.4 is better than previous models in Pro mode... crazy!

Coding capabilities are ridiculous... it's essentially flawless. Inside Codex, it's insanely reliable. Coding is essentially solved. There's not much more to say on this, it's just THAT good.

The Pro version is near-perfect. Other testers I spoke with saw it solving problems that were unsolvable by any other model. At this point, Pro is overkill for almost every normal use-case, but when you really need the power to do something extremely difficult, it's incredible.

Consistent with everything I've said above, even the standard thinking version uses fewer reasoning tokens than previous models to get the same level of results. In practice, this means you get great results much faster than before. This was one of my biggest gripes with previous OpenAI models. They just took too long to complete simple tasks. Assuming the speed we had during testing holds up as more users join, this is going to be a big win for OpenAI.

It still has weaknesses, though:

- Frontend taste is FAR behind Opus 4.6 and Gemini 3.1 Pro. , why is this so hard to fix? @OpenAI once you fix this, there's literally no reason for me to use any other model. Please please please do it!

- It can still miss obvious real-world context. For example, I had it plan an itinerary for a trip. At first glance, it looked perfect, but it failed to take into account that it chose locations that would be mobbed by spring breakers, so I had to re-run the prompt from scratch with more context.

- When testing it inside OpenClaw, it kept stopping short before finishing tasks. I'm assuming this will be fixed quickly, but it's still worth noting.

But zooming out:

This thing is so far ahead overall that the nitpicks are starting to feel beside the point.

GPT-5.4 is a serious fucking model.

The best model in the world.

By far.

[Amit Tripathi @ce_amit]

@iannuttall What if we use Anthropic skill for FE with GPT 5.4?

[Amit Tripathi @ce_amit]

Vibe coded project before going to Production

[Amit Tripathi @ce_amit]

Add this to 10× output quality in a coding agent:

Think simple and effective way embedded in SWE core principles and brainstorm with stress test to think that it would work in various edge cases.

[Amit Tripathi @ce_amit]

This is too good.

https://t.co/wN76MIveXc