Ihsan Engku

Im just an engineer who likes AI 🤖

[Ihsan Engku @DevEngku]

I Figured out the top 5 AI apps for losing weight

Watch the full video here https://t.co/UumsL5jkqb

[Tyler Shukert @dshukertjr]

With https://t.co/wd4jHo215I, you can create an unlimited number of databases using AI! You can create dummy data or import data from CSV files.

You can now deploy the database you built on https://t.co/wd4jHo215I to Supabase with a click of a button!

[Riley Brown @rileybrown]

Cursor Agent is just insane!

In a little over an hour, without writing one line of code I created a YouTube Super Analytics Tool and I filmed the whole process...

This is a full step by step guide on how to use Cursor Agent to build whatever you want with zero coding...

As always I used the @senior_swc nextjs template as a starting point that has been forked over 11,000 times. (in next tweet below)

Tools used:

- Cursor: Duh
- Replit: Template, web view and deployment
- Deepseek API: High quality AI Model (Cheap and better than gpt)
- SupaData API: Pull YT Transcripts
- SerpAPI: Search YouTube Results
- Firebase: DB and Auth

Chapters:

00:00 - YouTube Power Tool In Action
03:00 - Planning Phase (How I diagram and plan)
04:58 - Features I'm adding
06:20 - Setting Up APIs and Tools
10:00 - Developing the YouTube Transcript Extractor w/ Supadata + Deepseek
19:58 - The YouTube Search Feature (SerpAPI)
29:41 - Setting Up Google Firebase (DB + Auth)
32:00 - Implementing Search and Save Features
36:50 - Styling and UI Enhancements
49:00 - Final Result
53:40 - These features are being added to @yapthread but way better (First look at composer feature)

[Ihsan Engku @DevEngku]

Cool project

[Yohei @yoheinakajima]

introducing "ditto"
the simplest self-building coding agent

📄 ~500 lines of code
🛠️ can build multi-file apps
🔁 a simple LLM loop with 5 tools

github/replit/more below 👇

[william @wgussml]

🚀 I'm excited to announce the future of prompt engineering: 𝚎𝚕𝚕.

developed from ideas during my time at OpenAI, 𝚎𝚕𝚕 is light, functional lm programming library:

- automatic versioning & tracing
- rich local oss visualization tools
- multimodality native

Read on ⬇️

[Ihsan Engku @DevEngku]

This looks so clean

[Nico Albanese @nicoalbanese10]

this is all the code you need to create an ai agent with access to the web

using the vercel ai sdk, @ExaAILabs, and agentic (@transitive_bs)

[Gabriel Lespérance @GabLesperance]

@helloiamleonie https://t.co/2MVrwAXaZY for distributed compute
PyIceberg for my date lake
Next.js via v0 for the quick front end iterations
Vespa + Colpali for retrieval
Dspy for driving the complex LM programs

[Ihsan Engku @DevEngku]

Or even better learn python and generate your own algo for doing calculations

[Ihsan Engku @DevEngku]

LLMs like ChatGPT don’t count letters like humans—they rely on tokenization, This can cause errors, like the famous “strawberry problem,” where the LLM miscounted the “r”s. To avoid such issues, define the calculation process clearly or use a programming language.

[Ihsan Engku @DevEngku]

Or even better ask it to just generate the python code and you run it on your machine

[Ihsan Engku @DevEngku]

@vijayshekhar The only way to solve LLM strawberry problem

[Ihsan Engku @DevEngku]

Wow v0 looks so hot 🥵

[Lee Robinson @leerob]

Learn @v0 in under 10 minutes ✨ I'll show you how to:

• Build 3D games, run code, and create apps
• Learn about Next.js (with up-to-date knowledge of App Router)
• Start from @shadcn/ui components + fork from there
• Publish web apps and examples you can share with others

[Ihsan Engku @DevEngku]

I like this idea. AI and IOT automation feels like match made in heaven

[Jeremy @jrmyphlmn]

generative ui for home automations!

[Jeremy @jrmyphlmn]

generative ui for home automations!

[Ihsan Engku @DevEngku]

Wow this is good. But im wondering about the accuracy of this. Like how accurate it at handling different file formats

[Pontus Abrahamsson — oss/acc @pontusab]

Easily locate your company files using automatic classifications:

* Upload to @supabase
* Generate embeddings
* Structured output with @vercel AI SDK
* Real-time subscription

Next, hybrid search on embeddings 🔍

[Ihsan Engku @DevEngku]

Just spend 4 hours during my weekend building my own mindmap tooling. Inspired by @excalidraw
I have to say it is quite fun building. And the best part is the running cost of this is near zero. We’re talking about 0.001.
The site is here -> https://t.co/T6F1or9F0Z

[Ihsan Engku @DevEngku]

A good prompt use XML syntax

[Ihsan Engku @DevEngku]

Isn’t this the recipe of skynet?

[elvis @omarsar0]

Interesting new paper on automating the process of building powerful AI agents.

Presents Meta Agent Search a meta agent that iteratively programs and tests new agents based on a growing archive of previous discoveries.

Claims that with their approach it is possible to learn any possible agentic system including prompts, tool use, control flows, and more.

They achieve this by focusing on three main components referred to as search space (define agents), search algorithm (explore search space), and the evaluation function (evaluate candidate agents).

"We consistently observe the surprising result that agents invented by Meta Agent Search maintain superior performance even when transferred across domains and models, demonstrating their robustness and generality."

https://t.co/vPHls8Mxsq

[Ihsan Engku @DevEngku]

I noticed one trend here. If you want to capture the market you need to capture the heart of your users and Google is killing it!

[Logan Kilpatrick @OfficialLoganK]

Gemini 1.5 Flash free tier comes with:

- 15 RPM (requests per minute)
- 1 million TPM (tokens per minute)
- 1,500 RPD (requests per day)
- free context caching, up to 1 million tokens of storage per hour
- free fine-tuning

That’s 1.5 Billion tokens free, everyday.

(2/4)

[Ihsan Engku @DevEngku]

I love this take

[Dan McCarty @mccartydan]

@lemire speaking from experience (different api), this is accurate.

chatgpt will apparently reference api calls from other libraries or just create them out of thin air.

(kind of like a sales team!)

[Ihsan Engku @DevEngku]

This is one of architecture of your next js app should be for building highly scalable AI apps