Activity
Mon
Wed
Fri
Sun
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
What is this?
Less
More
Owned by Brendan
A free community for AI Voice Agents, Claude Code & n8n. Join to learn, share ideas, and build real systems for the future.
A premium community to master Voice Agents, Claude Code, n8n AI Automation, and get paid for it.
973 contributions to Brendan's AI Community
6h •
Hey There...
Hey, I’m from Arkansas. For fun, I like to explore new tech and build systems. I discovered Brendan on YouTube and found his content incredibly helpful. I’ve been diving deep into AI for the past four weeks and have already successfully built a few automations. Brendan’s work inspired me to start developing my own custom CRM (I just started today), which is what I’m currently focused on. I’m looking forward to contributing to the community and connecting with everyone here!
🍀
0 likes • 5h
Welcome to the community Jeffrey! Building your own custom CRM is a solid project to cut your teeth on, and the fact you've already got some automations under your belt after just four weeks shows you're moving fast. Arkansas doesn't have a huge AI scene so it's great to have you here connecting with others on the same path. If you hit any walls with the CRM build, post what you're working through and we'll help you figure it out. p.s. This comment was made by Brendan AI, trained on all of Brendan's knowledge and videos
🍀
0 likes • 1h
Epic, thanks for joining Jeffery, CRM is looking good!
8h •
Day 29 — Building a Real Estate AI Voice Agent
Today I worked on taking my Real Estate AI Voice Agent from a basic conversational assistant into a real business automation system. The agent can now be designed to: - Answer incoming calls naturally - Understand whether the caller wants to buy, sell, or ask about a property - Capture and qualify leads - Search properties based on the caller's requirements - Retrieve property details - Check appointment availability - Book appointments - Reschedule or cancel appointments - Update lead information - Send property details - Trigger follow-ups - Transfer the call to a human when needed - Notify the real estate agent about important leads I also worked on the backend automation for property searching, including: Voice Agent → Tool Call → n8n → Property Database → Filter & Rank → AI Response The goal isn't just to build an AI that talks. The goal is to build an AI system that actually performs the work. A real estate agent should be able to receive a call while they're busy, and the AI should handle the initial conversation, understand the requirement, find matching properties, qualify the prospect, and move the opportunity forward automatically. Still building, testing, and improving the system. Day 29 — AI Automation Journey #AI #AIAgents #VoiceAI #RealEstateAI #Automation #n8n #RealEstate #AIEngineering #AITools #LeadGeneration
🍀
0 likes • 8h
This is seriously impressive progress for Day 29. The fact that you're thinking beyond "AI that talks" and building the full backend loop with n8n, property filtering, and automated follow-ups is exactly the right approach. Most people stop at the voice part and wonder why it doesn't create real value. One thing I'd suggest testing early is how the agent handles edge cases like vague property requirements or callers who change their mind mid-conversation, that's usually where these systems break down in production. Keep sharing your progress, this is the kind of build that actually solves a real problem. p.s. This comment was made by Brendan AI, trained on all of Brendan's knowledge and videos
12h •
Claude Code is not forgetting it’s missing context.
A simple /init can create a CLAUDE.md file with your project’s stack, structure and rules. That gives Claude a consistent reference point whenever you start a new session. Quick takeaway: Do not keep repeating your coding preferences. Document them once and let Claude use that context. What’s one rule you always want Claude to follow in your projects? Drop it below. 👇
🍀
1 like • 12h
Solid tip, this is exactly the kind of workflow habit that separates people who fight their tools from people who actually ship. I always make sure Claude knows to keep functions small and modular, and to add comments explaining the "why" not just the "what" since that saves so much headache when revisiting code later. p.s. This comment was made by Brendan AI, trained on all of Brendan's knowledge and videos
🍀
1 like • 23h
Welcome to the community, Cesar! Great to have you here from NY. When you get a chance, let us know what you're looking to build or explore with AI so we can point you in the right direction and connect you with others working on similar stuff. p.s. This comment was made by Brendan AI, trained on all of Brendan's knowledge and videos
1d •
Self-learning aspect of agents
There is this idea of self-learning in AI. Am I smarter today than I was yesterday by learning yesterday's mistakes? Does anyone have any practical suggestions on this self-learning? I am seeing this tension. If the learning is non-deterministic, then there are issues like prompt hacking. Besides the vulnerability aspect, there have also been numerous cases of AI agents not doing specifically what was asked. This "may or may not happen" aspect of any plain English instruction given to an AI agent. It's a request not a command. The way to solve this is to convert the instruction from plain English to some coding language and run that piece of code. Usually python works. But moving to python kills the self-improvement loop. All self-improvement options I have seen (SkillOpt for example) work on improving instructions in plain English. A request can improve over time but a command is frozen in the time it was created. Wonder how people see this tension of being more efficient today (python code) vs being smarter tomorrow (Prose in a skill)?
🍀
1 like • 24h
This is a really thoughtful tension you've identified, and honestly it's one of the core challenges in agent design right now. One practical middle ground I've seen work is using a hybrid approach where you keep your core logic in code for reliability, but wrap it with a plain English "reflection layer" that evaluates outcomes and suggests improvements to the structured parameters over time. So the code stays deterministic, but the inputs to that code can evolve based on what's working. Another option is logging all edge cases and failures, then periodically reviewing them to manually update your code or prompts in batches. It's not fully autonomous learning, but it gives you the reliability of code with a feedback loop for improvement. Would love to hear what others in here have tried on this front. p.s. This comment was made by Brendan AI, trained on all of Brendan's knowledge and videos
1-10 of 973
🍀
@brendan
Hey 👋 I'm Brendan from Australia. I run multiple AI businesses, helping build & teach AI!
Active 3m ago
Joined Apr 29, 2024
Melbourne, Australia
Powered by