The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
ML pipeline trained on 6.3M real-world transactions with extreme class imbalance (~0.13% fraud rate).
The hard part wasn't the model — it was tuning decision thresholds against actual business metrics (precision-recall tradeoffs, cost of false negatives vs false positives) rather than chasing default accuracy numbers. Deployed as a live REST API via Flask for real-time inference.
Stack: Python · XGBoost · Flask · Scikit-learn · Pandas
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Trending
retrofuturism
Share your favorite designs from the past exploring the distant future. Or your own take on the aesthetic 🚀
halftone
Dots. Texture. Grain. Let's see your designs using the halftone technique.
Midjourney
Creatives are showing off what they're making with Midjourney.
colorgrade
Video editors sound off: How do you enhance the mood of your video in color grading?
Related posts
Beautiful UI shouldn't require messy code. ✨
A while ago, I built this feature section for TipMi. Looking at the varied card sizes, you might think it requires complex positioning or endless Flexbox containers. The secret weapon? CSS Grid with grid-template-areas.
Using Grid for this specific architecture is the easiest, most scalable approach. You get complete two-dimensional control, allowing you to span items across multiple columns and rows seamlessly while keeping the underlying HTML structure incredibly clean and readable.
Take a look at the live build here: https://tipmi-dev.vercel.app/
Great design 👍
Trading Analytics & Self-Learning System
A data-driven analytics and self-learning module built for an automated Telegram trading system.
The system continuously records trading signals and completed trades, analyzes historical performance, and provides real-time statistics directly through Telegram.
It tracks win rate, total and average PnL, best and worst trades, signal distribution, confidence scores, and trading performance over time.
Key features:
Automated trade history tracking
Real-time PnL calculation
Win rate monitoring
BUY / SELL / HOLD signal statistics
Confidence score analysis
Historical performance evaluation
PostgreSQL data storage
Telegram reporting interface
Automated signal logging
This module was developed as part of a larger Python-based automated market analysis and demo trading platform.
Automated Trading Signal Analytics
A real-time analytics module for monitoring and evaluating automated cryptocurrency trading signals.
The system automatically stores and analyzes generated BUY, SELL, and HOLD signals, tracks confidence scores, records market prices, and provides performance statistics directly through a Telegram interface.
Key features:
Automated BUY / SELL / HOLD signal tracking
Real-time signal statistics
Confidence score monitoring
Historical signal logging
Trade performance and PnL analytics
Win rate tracking
PostgreSQL data storage
Telegram reporting interface
Automated background monitoring
This analytics module is part of a larger Python-based automated market analysis and demo trading system.
