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
SQLEval is an AI evaluation platform designed to benchmark and compare the accuracy and reliability of LLM-generated SQL. It runs controlled evaluations across different models and prompts, measures performance, and helps identify failure cases through detailed results and analysis.
Built with Python and modern LLM evaluation workflows, SQLEval focuses on reproducible testing, model comparison, and understanding how AI systems perform on real-world SQL tasks.
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
One of my heavy CMS automation work inside Framer (A year ago...)
The client wanted a simple way to manage new renovation leads and make sure no inquiry was missed.
I set up a HighLevel workflow that collects new leads, checks their budget, sorts them by priority, and sends a confirmation email right away.
The client also wanted follow up emails to stop when a lead replied, so I added that as well
After setting everything up, the client had a simple system that handled new leads automatically. Leads were captured, sorted by priority, contacted right away, and followed up without the team having to do it manually.
This helped the client save time, stay on top of new inquiries, and make sure good leads did not get missed.
The reply detection stopping the follow up sequence is the detail that separates a well built workflow from one that annoys leads into going cold. most people set up the follow up and forget to add the exit condition. that one addition alone changes how the whole system feels to the person on the receiving end
Built this fully automated AI video pipeline — from script to finished, published video, zero manual editing.
I mapped out and built an end-to-end n8n workflow that takes a topic and turns it into a complete video: AI writes the script, generates the voiceover, creates the visuals, auto-syncs captions, assembles everything, then stores and logs it automatically.
The hardest part wasn't any single step — it was getting all the tools to talk to each other cleanly. Each stage had to pass the right data to the next without breaking, so one failed generation wouldn't quietly kill the whole run. Lots of trial, error, and rebuilding connections until it just... worked.
End result: a system that can produce content on autopilot, at a scale a person doing it manually never could.
