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
Built a voice agent for vendor follow-ups and customer ops on OpenAI’s GPT-Live-1 API. In a small test with 4 users, it felt instant and natural.
That’s not the same as ready for a client.
Before I’d hand something like this over (or open it to more users), I want to see:
→ Speed holding up as concurrent calls increase
→ Interruptions handled cleanly on real phone audio
→ Names, IDs and phone numbers captured correctly
→ Calls that end properly, plus graceful fallbacks when something times out or drops
→ Real cost per minute, including the backend model
A demo that works and a product that holds up are different jobs. I’m working on closing that gap now.
If you’ve shipped a voice agent for a client, what broke first?
Open to a few voice-agent projects this quarter
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
Related posts
Exploring the listening state for an AI voice interface.
Which direction would you ship, A or B?
10 voted
29%
25 voted
71%
35 votes
Closed
The particle field looks more polished and beautiful🔥
This year’s YC Fall batch is for us creatives. 🎯
As a product designer, I’ve never felt like our field was changing as rapidly as it is right now. Tools emerge, you adapt, they disappear. Processes that took hours can now be largely automated, and overall, I feel like a big part of the fun gets lost because of a communication issue: I struggle to tell the tools exactly what I want in enough detail.
I recently stumbled across a few interesting and promising startups that might bring some of that joy back and take the pain out of parts of our work.
Proto is building an IDE specifically for product designers, bringing design and AI workflows closer together. 👩🏻💻
Simulithic uses AI agents to test your app across devices, browsers, and account types, automating test creation and maintenance. 🤳🏼
Composal uses swarms of AI agents to explore product flows and edge cases, finding what breaks before release. 🕵🏻♂️
Native helps marketers decide what to create by ranking ideas based on brand context, performance, trends, and creative feedback. 🦖
The edge case testing sounds promising.
An AI customer support chatbot for a fictional Nigerian food delivery business, Lagos Fresh Kitchen. It runs on n8n Cloud and uses Claude Haiku 5.5 as the AI model. The workflow is three nodes: a chat trigger receives the customer’s message, an AI Agent processes it, and the Anthropic Chat Model writes the answer.
How it behaves
The AI Agent has a system message that sets the rules. It must answer only from the business information, never invent prices, areas, hours or policies, and use a fixed fallback line when the answer isn’t there: “I don’t have enough information to answer that accurately. Please contact our support team.”
The business information it knows
Open Monday to Saturday, 9 AM to 8 PM, closed Sundays
Delivers to Lagos Mainland (Ikeja, Yaba, Surulere, Maryland, Ojota, Gbagada) and Island areas (Victoria Island, Ikoyi, Lekki), not outside Lagos
Delivery takes 30 to 60 minutes, with fees from ₦1,000
Menu of jollof rice, fried rice, rice and stew, pounded yam with egusi, grilled chicken, suya and soft drinks
Payments by card, transfer and cash on delivery
Cancellation allowed before cooking starts
No Sunday orders or Sunday delivery, and no event catering
Refunds, complaints and order status go to support
good work