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
In this project, I developed an automated solution that fetches Twilio call recordings, transcribes them using OpenAI API, and stores the transcription as a note in Go High Level (GHL). The solution leverages n8n workflows for seamless integration, automating the entire process from recording retrieval to transcription and note creation.
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
62% of business calls to small businesses go unanswered. I just shipped the fix.
Every missed call is a lead one tap away from calling your competitor instead. So I built an n8n system that closes that gap automatically:
⚡ The instant a call is missed, it looks the caller up in your CRM
🤖 AI drafts a personalized recovery SMS — not a generic "sorry we missed you"
🧠 If they reply, AI reads it and tells a hot lead from a cold one
📞 Hot leads get escalated straight to a callback task, assigned to your team, in seconds
📋 Every touchpoint logs to your CRM automatically — no spreadsheets, no gaps
16 nodes. Two entry points. Built on n8n + Twilio + JustCall + OpenAI, wired for Less Annoying CRM (or repoint the CRM steps at whatever you run — they're plain HTTP requests).
No monthly subscription. No vendor lock-in. Buy the template once, connect four credentials, and it's running on your own infrastructure — not another SaaS charging you forever for your own data.
Live on Contra now — $59, with setup notes built right into the canvas and a full README walking you through every credential.
📩 DM me if you want this customized for your exact stack.
#Automation #n8n #AIAutomation #Twilio #NoCode #LeadGen #SalesOps #Contra #Freelance
Pocketride is a ride-hailing app for riders and drivers - book a ride, get matched with a driver, get where you're going. We built it for mid-size Canadian cities like Sarnia, which are too small for the big platforms (Uber etc.) to prioritize but have plenty of real demand. It's currently live with real drivers and riders.
How we built it on Bubble: Server-side recursive dispatch workflows -moved all ride-matching logic off the client so it keeps running even when a driver's phone is minimized or backgrounded, fixing a major reliability gap from client-side timers.
Push notifications via the Devices field - needed for push to reliably reach drivers across app states.
Deep-linked navigation to Google Maps - native mobile doesn't support HTML elements, so we route drivers out to Maps for turn-by-turn instead of building nav in-app.
Twilio SMS as a redundant channel alongside push, so notifications land even when FCM delivery is inconsistent. And, also for phone OTPs while signing up.
Stripe for payments, with QuickBooks Contractor Payments integration in progress for direct driver payouts.
A single shared ride-claim action to prevent race conditions when multiple drivers get notified about the same ride at once.
The fun part of building this on Bubble was proving out a real-time, race-condition-sensitive, notification-heavy system, the kind you'd normally assume needs a custom backend, entirely on the platform.
Amazing work wish you the best!
This portfolio sample shows how I evaluate AI agents and automated workflows to find reliability problems, broken logic, and weak edge-case handling. I review whether the system completes the intended task, uses tools correctly, follows instructions, handles ambiguity, and produces an output that can actually be trusted.
The work focuses on identifying failure points before they become real user problems, including incomplete task execution, incorrect assumptions, workflow dead ends, hallucinations, and verification gaps. This is a portfolio demonstration of my AI agent evaluation and red-teaming work. No confidential client information is included.
Thorough breakdown! Robust edge-case handling and reliability checks are definitely the trickiest part of building reliable AI agents 🤖📊