A lot of people from here shared their experiences with me in my last post, when I asked about the problem of self hosting n8n within a VPS. Hetzner and Railways stood out as 2 incredible options to me and I went ahead with hetzner and today, I can literally create events from my phone, without my laptop running on any quick tunnel or cloudflare or etc. and I am just so happy that I have done what I really thought would take me longer than this to achieve! I can now add people to my config that I am using right now, and ask their experience on event creation with my whatsapp bot. Too happy :) Please share any leads you have for any small businesses requiring this kind of aid. I am happy to connect with any person, arrange a meeting or answer any of the question your network brings! Onwards and upwards!! 🔥 🚀🚀

Hello new here. Hoping to learn about AI automation and become a part of the community

Hey everyone, Sent this to the list late last week, posting here in case you missed it. We went back through Q3, 1,150+ agencies, and pulled the 10 biggest things separating the ones winning clients from everyone else. 1. Demos beat proposals. Too many agencies spend hours explaining what they could build. The better ones let the client see it. Thomas won a $20,000 healthcare project (search and drafting across 13,500 regulatory documents, plus a chat agent and a voice agent). Other firms pitched their systems. Thomas let the client use his. The client is no longer imagining whether you can deliver. (see #2 attached image) Next serious opportunity: build a small demo around one real piece of their workflow before you send another 20-page proposal. 2. Diagnostics beat discovery calls. Get paid to understand the problem first. Stephanie sold a $3,000 diagnostic to a health-tech startup one week after joining, a broken funnel across three products. (see #3 attached image) Instead of pitching software, she got paid to work out what was broken and why. Easier first purchase for the client, far better position for you to sell the implementation. Diagnose, put a number on it, then recommend. 3. In-person beats hiding behind your screen. Briland closed three clients in one week after meeting owners at in-person events. Three clients, three industries, one channel. (see #4 attached image) There are owners in your city at breakfasts, conferences and networking groups right now talking openly about problems they need solved. Find one room your ideal clients are already in and go this month. 4. Warm beats starting from zero. In Q3, 38% of new deals inside the Accelerator came from warm relationships: past clients, old colleagues, partners, friends. Yet owners ignore their network because cold outreach feels more scalable. Your warm network already has the hardest thing to manufacture: trust. Before another 500 cold emails, list 30 people who already know you.

Been sick all day tried to thug it out, couldn't. Hopefully ill be able to things done tomorrow. If anyone has some secret sauce on how to get through being sick I would appreciate it a lot!

A few weeks ago I built a fairly complex automation with Next.js on the frontend and Python handling most of the backend logic. Everything looked fine. Tests were passing. Logs looked normal. The automation was doing what it was supposed to do. Then one night around 2 AM, the LLM returned malformed JSON. Normally you'd expect something to fail loudly. This one didn't. Part of the workflow continued, part of it stopped, and the failure wasn't obvious until the client noticed something was wrong. That was the uncomfortable part. The system hadn't really failed. It had failed silently. The client found it before I did. I was honestly lucky they did, because this was a high-value client and it could have turned into a very different conversation. I started looking at how I was monitoring the automation and realised something I hadn't really thought about before: Most of my tooling was very good at telling me what happened. But I needed something that could answer: Did the automation actually produce the outcome it was supposed to produce? So I started building a layer around the automation that doesn't just watch errors. It checks the output against the expected structure, looks for suspicious states, catches things like malformed responses, retries and rate-limit behaviour, records the evidence, and tries to determine whether the workflow should actually be considered successful. The interesting part is that I've found quite a few cases where: HTTP 200 workflow completed LLM said "success" …still didn't mean the business outcome actually happened. I'm about 60% through rebuilding this properly, and I'm beginning to think the difficult part of AI automation isn't getting the workflow to run. It's proving that it ran correctly. Now I'm going down a bit of a rabbit hole with failure reproduction and regression testing too, because fixing a failure once doesn't tell you whether the same thing will happen again three weeks later. Curious how other people are handling this.