Crypto Trade-R
See crypto hype before the price moves.Our dashboard makes it simple, clear, and easy for anyone to use.
Join the waitlist for early access.
Blog Posts
Follow along as I build
Crypto Trade-R, sharing progress, lessons and early ideas.
Behind the Radar:
Notes from my journey of building Crypto Trade-R.
Can We Spot Crypto Hype Before the Price Moves?
My journey building CTR + Cortex (so far)
By Thina Chelvam - September 21, 2025
Can We Spot Crypto Hype Before the Price Moves?
My Journey Building CTR + Cortex
By Thina Chelvam - September 21, 2025
When I first started playing with this idea, I had one simple (and slightly ridiculous) question stuck in my head:Can we spot crypto hype before the price moves?It sounded obvious enough, markets move on sentiment, traders live on Twitter/X, Reddit never sleeps, surely you could measure the chatter and see the pump coming? Right?Well… not exactly.
The Scrappy Start: Google Sheets as a “Database”
Like any sane person trying to solve crypto sentiment, I opened up… Google Sheets.Yep. I hacked together some code, tried pulling in posts from social sites, and convinced myself I’d built the foundation of the next Bloomberg Terminal.What I actually built was:1. A shrine to error messages.2. A graveyard of half-working API calls.3. A spreadsheet that looked more like spaghetti than usable sentiment data.I restarted the project maybe 15 times. Every time I thought I’d cracked it, something else broke:- API rate limits.- Bots shouting “XRP to $10,000!!!” on repeat.- Data that made zero sense once you actually looked at it.At one point, I was literally scraping posts into Sheets and wondering why it felt like duct-taping the internet to a calculator.But I kept going.
Lessons in the Chaos
I’ll be honest: most of the time I didn’t really know what I was doing.I was impatient, frustrated, and had no clue how best to analyze what I was pulling. Half the time I wasn’t even sure what I should be analyzing.But those messy experiments taught me a lot:- Data integrity matters. If you don’t clean and filter properly, you’re just charting noise.- Bots are everywhere. If your system can’t handle that, it’s useless.- Free tools are gold. Google Sheets and free APIs got me far enough to prove something was possible.And most importantly: hacking was fun, but if I wanted to build something real, I needed to step back and design it properly.
The Turning Point
That’s when I asked myself: What’s the actual end goal here?It wasn’t just to pull random data into a sheet. It was to help traders see real sentiment clearly, hype vs credibility, signal vs noise.That meant building something more ambitious. Not just a dashboard. Not just a spreadsheet.I realized I needed to build… a brain.
Enter CTR + Cortex
That’s where things started to click.- Crypto Trade-R (R = Radar): The dashboard I want to build, clean, retail-friendly, no clutter, no overload, just clear signals.- Cortex: The AI “brain” I’m designing to sit behind it, something that learns, filters the hype and shows what’s real.Institutions already use advanced sentiment platforms. Retail tools exist, but most are noisy, costly, or hard to trust. CTR + Cortex is about bringing clarity and accessibility.
The Journey Ahead
So that’s where I’m at now. I’ve gone from messy Google Sheets experiments to a much clearer vision of what CTR and Cortex should become.This time around, I’m starting fresh, properly designing the foundations, instead of duct-taping APIs into a spreadsheet.Over the next months, I’ll be:- Building CTR into its first live dashboard.- Training Cortex to get smarter at filtering out noise and bots.- Sharing weekly progress updates, the wins, the fails, and the lessons.If you’re curious to follow along (or want early access once the first version is live), you can join here: (CTR waitlist.)This is just the beginning. The question that kicked it all off still drives me:Can we spot crypto hype before the price moves?I don’t have the answer yet, but I’m building towards it.
Not financial advice. Just one person, a lot of coffee, and way too many error messages.