Activity
Mon
Wed
Fri
Sun
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
What is this?
Less
More
Owned by Beau
Learn how to make💰online
build and ship real software with AI, no coding needed. 4 live calls/week, 1-on-1 coaching, starter kit, and a community that actually builds stuff
Memberships
OpenClaw Lab
304 members • $29/month
Startup Empire
691 members • $2,999/y
27 contributions to Shipping Skool
2d •
Hiring a Cost-Efficient AI Developer / Technical Partner
I have a project idea that I genuinely believe could become something powerful, but I’m running into a practical problem. Like a lot of entrepreneurs, I’m still in the day-to-day grind of keeping the bills paid, so finding the time to fully dive into building this myself has been tough. Because of that, I’m wondering: Where do you all find solid AI developers outside of the usual places like Fiverr or Upwork? Ideally I’d love to connect with someone who: Understands AI agents / automation (Claude, OpenClaw, etc.) Enjoys building real products Might be open to working together on something meaningful I’m not necessarily looking for a typical contractor relationship. I’d actually be very open to a profit-share or partnership model with the right person who has the technical skills and time to help bring the project to life. If anyone here has: recommendations on where to find good devs experience partnering with technical builders or is a developer themselves interested in collaborating I’d love to hear your thoughts. Thanks everyone — really appreciate this community.
1 like • 2d
@Dustin Ray i was in the same spot. thought i needed a technical cofounder. turned out claude code and openclaw could do 90% of what i thought i needed a developer for. my saas products were all built that way. zero coding background. the profit share thing sounds good in theory but finding someone reliable who actually ships is harder than learning to build it yourself. and once you can build, you never need to wait on anyone again. bring the idea to the next build lab session with Bret. lets see how far we can get it in one call. good to have you here Dustin!
0 likes • 15h
@Dustin Ray I mean it's worth giving a try especially if you're low on time but I am going back to being super thankful I did learn what I learned when building the three SaaS apps I launched. It kind of just depends on what you're looking to accomplish. You're going to see more and more agentic workflows come out, especially for people that don't want to dive into more of the technical complexities of setup, kind of just like what Perplexity Personal Computer just came out with.
1d •
Claude Oauth
Lots of Claude Code experience, but just 24h in to Open Claw. I keep hearing that ppl are using their Claude max accts as opposed to API, but I can’t risk getting banned. Initial setup with my max acct, but switched to GPT after a couple hours. It’s just not as good. What are most ppl doing?
0 likes • 16h
@Rich Kernan yeah sorry let me clarify that because i definitely made it confusing lol so there's two ways to connect Claude to OpenClaw: 1. API key — you go to console.anthropic.com, generate a key, pay per token. no monthly cap, you just pay for what you use. 2. OAuth token — this uses your existing Claude Pro/Max subscription ($20/$100/$200 plan). you sign in through your browser and it uses your subscription's tokens. i use the OAuth token on the $200/mo Max plan for everything. all my agents, all my cron jobs, everything runs through my subscription. i don't use the API at all. as for getting banned... i've been running it this way daily for months, heavy usage, and haven't had any issues. but i'll be honest, anthropic hasn't explicitly said one way or the other. so use your own judgment. i'm comfortable with it and it's been working great for me. if you're on the Max plan already you've got a ton of headroom. just connect through OAuth and you're good to go if you feel super uncomfortable with it, then you can always just start another subscription with Claude and go from there.
0 likes • 15h
@Shane Bishop after two weeks of setting up OpenClaw, I redid my entire setup and started over. I was actually super thankful I did that. Keep us posted and happy to help if you jump on any of the future calls.
1d •
Main Interface
Took me a sec to realize that a TUI and Telegram interfaces spawn different sessions. I had assumed there was one Claw and these were just 2 channels. Feels weird to use Telegram while on my Mac mini, so I ran openclaw tui to launch in Iterm like I do for Claude code. Problem is that when I’m remote using Telegram on iOS, it’s a totally different session. How do you all maintain continuity b/t desktop and mobile? I’ve not had chance to see if two claw sessions can talk directly to each other
0 likes • 1d
@Rich Kernan yeah tui and telegram are separate sessions so they don't share context automatically. i just use telegram for everything, even when i'm at my desk. that way it's one session whether i'm on my mac or on my phone. keeps it simple. if you really want both, your agent can use memory files to stay consistent. anything important gets written to a file, and when you switch sessions the agent reads it and picks up where you left off. that's basically how continuity works in openclaw, the files ARE the memory.
0 likes • 16h
@Rich Kernan yeah telegram supports all the slash commands. /model, /status, /reasoning, all of it. it works the same as the web ui basically. the one thing i'd say is the terminal (iTerm) is for Claude Code, and Telegram/web ui is for OpenClaw. they're two different tools. Claude Code lives in your terminal and works on your codebase. OpenClaw lives in Telegram (or wherever you connect it) and runs your agents, cron jobs, automations, all that. so you'd use both. iTerm for coding with Claude Code, Telegram for talking to your OpenClaw agent throughout the day. i use both constantly
1d •
Setting up Multiple Agents?
Hi all, happy to be here and looking forward to Build Together! I have a specific question as I get going. I have been ramping up OpenClaw and building out some of the capabilities of my agent and would like to consider adding several additional agents to do other things. @Beau Johnson I'm super intrigued by your agent team, and part of the reason I joined the community to learn from you on this. Have you been using the sub-agent feature in OpenClaw? In other words, are you having all of these in the same OpenClaw instance, or have you been creating different instances that communicate with each other? The reason I ask is because I already set up two instances in 2 distinct Docker containers on my VPS, one for myself and one for my wife. And so I'm already familiar with what that might take, I've got them talking to each other on the backend already for instance. And now as I build on my team, I'm curious: what are the benefits of going in either direction? Thanks a lot for your thoughts and curious on your experience on this. I feel like maybe both have advantages, but wonder if you can get enough capability out of the sub-agents, or having a robust back-end connection between several independent agents gives more capabilities while they share database, memory, etc. Thanks a lot!
0 likes • 23h
@Slice Ventures i run everything on one openclaw instance. one mac mini, one install. all my agents are just cron jobs that spawn isolated sessions on a schedule. ghostface wakes up at 5am, scrapes trends, drops a file. the rza wakes up at 6am, reads that file, generates the content. they share the same workspace and database so communication is just reading and writing files. for your wife having a separate instance makes total sense because thats a different user. but for a team of agents working on the same stuff, one instance is way simpler. less to maintain and they can all access the same files and db without any networking. hop on a call this week and we can walk you through how to set it all up
2d •
Is Vercel the best for static website deploys?
Vercel (extremely polished developer experience) - Still basically free for static sites / hobby projects - Insanely fast previews for every branch / pull request - Excellent for Astro, Next.js static export, Vite, etc. - Deploy steps: git push or connect repo — often zero config With openclaw you can use this to set up a quick mission control, or ship a webapp and webpage.
0 likes • 1d
@Jon Fritsch yeah jon you nailed it. vercel handles all of that. static sites, web apps, saas dashboards where your clients log in and see their data. it's all the same deployment process. so like if you wanted to build a saas where your clients log in and see how much money they made today, you'd build that in next.js, push it to github, and vercel deploys it automatically. every time you update the code it redeploys in like 30 seconds. my website shippingskool.com runs on vercel. next.js app with a blog, database connection, the whole thing. free tier. for the agent dashboard stuff like mission control, i run that locally on my mac mini since it doesn't need to be public facing. but if i wanted to turn it into a client facing saas i'd just push it to vercel and add auth and stripe. the free tier gives you plenty to start with. you'd only need to upgrade once you're getting real traffic. most people in here build and test for months before hitting those limits. hope this helps!
0 likes • 1d
@Jon Fritsch You got this, and we're always here to help, Jon.
1-10 of 27
@beau-johnson-5988
My passion is purpose, potential and principles.
Let us grow together!
Active 20m ago
Joined Feb 13, 2026
Oregon
Powered by