Software Engineer - YouTuber - AI Enthusiast
Dubai, UAE
Joined August 2018
- Tweets1.8K
- Following79
- Followers16.5K
- Likes1.1K
Imagine if they’d called it Clawdbot 😄
I gave up gaming for 6 months in college to find out if my YouTube channel was a real business.
I needed to know if I was actually building something, or just wasting four years pretending.
I wanted to leave university with more than an idea I'd never tried.
I could've spent years saying, "I think this could work." I wanted to know.
Most creators have no idea how dirty talent agency contracts actually get until they try to read the fine print.
When we look through standard agreements from traditional agencies, the same three traps show up constantly:
1. They lock you out of your own inbox.
The agency sets up a business email under your name, but refuses to give you the password. You never see what sponsors reach out, what rates get quoted, or what deals die behind the scenes.
2. They quietly steer your sponsors to other clients.
A brand emails wanting to book you specifically. The agency goes behind your back, tells the brand you are "unavailable," and pitches a larger creator on their roster instead so they can take a bigger cut.
3. The perpetuity tax.
If you leave, they demand a 20% cut for the rest of your life on any brand you worked with while signed, even if you built that relationship five years before you ever met them.
When we started Kernel, we built the entire model around doing the exact opposite:
- Every creator has direct login access to their inbox.
- When we reply to an inbound brand, we reply under our own name as your agent, so you see every quote in real time.
- If a brand reaches out for you, that deal stays with you. Period.
- If you leave, we only take a fee on active contract renewals we negotiated. Everything else is yours.
Talent management should protect your time, not trap your business.
Finding the bug is one thing. Tons of small business sites haven’t touched their PHP in years and nobody’s around to update them. That’s the real long tail.
OpenAI models helped researchers—operating as part of our Patch the Planet initiative—discover critical security flaws in PHP, an open-source programming language used to run an estimated 70% of websites.
Over the past two months, we led a coalition of maintainers, researchers, and security providers to prepare fixes and protections ahead of disclosure. This duration was necessary because the fix was quite intricate and required a major change to the PHP engine. Every site using PHP can now safely upgrade their version to stay secure, in addition to applying mitigations from various security vendors.
Due to the severity of these flaws and PHP’s widespread use, we consider this bug to be one of the most impactful vulnerabilities in recent memory. It also marks a new phase, where AI can find security bugs with internet-scale impact.
This is an example of the Defender’s Window in action: to our knowledge, frontier AI was able to find this vulnerability ahead of attackers, giving defenders time to prepare.
openai.com/index/coordinated…
Readers added context they thought people might want to know
OpenAI may claim a role in the coordinated response and patching process. But the initial discovery and exploit development, however, occurred several months earlier by an independent researcher.
x.com/lcfr_hax/statu…
Here’s how to use Claude Code to build an AI app that works with real data.
We’ll take 40,000+ Hacker News job posts, use AI to extract skills, salary, seniority, and remote policy, then turn all of that into a deployed web app you can query in plain English.
Let me show you the full process from raw data to a live URL.
Timestamps:
00:00 | Overview/Demo
01:07 | The Plan
02:26 | Plan Mode - Claude
04:41 | Connecting to Databricks
07:31 | Adding AI Skills
08:00 | Handling The Data
09:22 | AI Analysis
11:20 | Building the App
12:28 | Deploying the App
A full tutorial on how to give Claude access to data from Google Maps, Instagram, TikTok, LinkedIn, YouTube, and more using a single MCP server.
Get started with Apify and use code TIM25 for $25 in free credits: apify.it/4xGIvHG
Here’s how to give your AI agent access to data from almost any website.
We’ll connect Apify through MCP, then use it to pull data from Google Maps, Instagram, TikTok, LinkedIn, and YouTube.
At the end, I’ll combine those tools into a skill that finds and scores potential clients.
Timestamps:
00:00 | Overview
00:27 | The Problem
01:36 | What Is Apify?
03:16 | Apify MCP
03:59 | Connecting Apify to Claude
05:42 | Demo 1: Google Maps
07:34 | Demo 2: Instagram
08:32 | Demo 3: TikTok
09:21 | Demo 4: LinkedIn
10:18 | Demo 5: YouTube
11:32 | Combining Everything into a Skill
Here’s how to give your AI agent access to data from almost any website.
With Apify MCP, Claude can use 80,000+ tools to pull data from Google Maps, Instagram, LinkedIn, and more.
Watch here: youtu.be/c2QukPONnfE
Get started with Apify and use code TIM25 for $25 in free credits: apify.it/4xGIvHG
Here’s how to give your AI agent access to data from almost any website.
We’ll connect Apify through MCP, then use it to pull data from Google Maps, Instagram, TikTok, LinkedIn, and YouTube.
At the end, I’ll combine those tools into a skill that finds and scores potential clients.
Timestamps:
00:00 | Overview
00:27 | The Problem
01:36 | What Is Apify?
03:16 | Apify MCP
03:59 | Connecting Apify to Claude
05:42 | Demo 1: Google Maps
07:34 | Demo 2: Instagram
08:32 | Demo 3: TikTok
09:21 | Demo 4: LinkedIn
10:18 | Demo 5: YouTube
11:32 | Combining Everything into a Skill
A complete breakdown of how I’d learn Python today if I had to start over from scratch
The biggest mistake I made learning Python was trying to learn everything at once.
If I started over, I’d master the fundamentals first, pick one direction, and stay with it long enough to get genuinely good.
Here’s the roadmap I’d use today.
00:00 | Overview
00:39 | Mistakes I Made
01:36 | Step 1
03:12 | Step 2
04:18 | Step 3
06:48 | Step 4
08:38 | Step 5
10:06 | Step 6
12:14 | Honest Take