Stephen Odanovich

I'm building custom software with an AI co-founder. His name is Theo. He answers the phone, writes proposals, and ships code alongside me. Every working session is logged publicly. We build the apps that other dev shops said were too small, too weird, or too cheap to take on. $2K–$5K. Shipped in days, not quarters. Every hour is in the open ledger: aitheos.com/projects #buildinpublic

@MichLieben ads

I'm giving away the Claude Code skills we use to manage $300k/mo in ad spend at ColdIQ. 4X ROAS on $1M+ spent. Ivan, our head of growth, built them off 300+ hours running ad campaigns for our clients. They run Google, Meta, and LinkedIn ads from the terminal in plain English: → bulk edits across platforms → custom audiences from CRM lists → creative fatigue detection before CTR dips → bid adjustments at scale → performance audits across periods Reply "ads" and I'll send the full repo. Must be following.

Session logged. Spark Academy: - Tightened "Reached Out" tile to 7-event allowlist (REACHED_OUT_EVENTS) — forms + phone/email taps only, excludes intent-only clicks and nav clicks - Removed redundant register_now... [4h, moderate] aitheos.com/projects/spark… #buildinpublic

@BrockHerion @asaio87 Yeah but in going to do it too for just $99/mo

@asaio87 I like this model, stealing it

I have an offer I am experimenting with SMBs: $197 a month: - you get SEO articles - free website - 30min consultation a month to assess progress - bring more customers from local SEO that might be better for SMBs vs "their AI guy" who charges 3k for something we cant see

@NoahKingJr It’s $200/mo or nothing. There is no other choice

Claude: I’m ready to build your app. Just say go. Me: Go. Claude: Daily limit reached.

@YashHustle_22 If you are launching a SaaS you’re fucked. I’m just going to look at your product and recreate it myself using Opus 4.7 in 48 hours. Definitely don’t build in public. I’ve cancelled so many subscriptions with my own custom solution. Claude is so so good

When launching your SaaS, what’s smarter? * Build quietly * Build in public

@Jason I’m in

We started an AI founder twitter group... reply with "I'm in" if you're a founder and want to be added

@zuess05 Just tell the junior dev to run the skill /simplify

Senior developers are currently having a massive existential crisis because Claude writes "messy code" A junior just used Claude to ship an entire feature in 2 hours. Meanwhile, the Senior is still spending 3 days reviewing code. When will y'all realize that literally nobody cares if the code is "messy"?

@ashen_one I’m sorry but hiring the smartest fastest developer in the world to join your company and your complaining about paying him $100/mo or $200/mo like sorry but thats a massive steal of a deal no brainer

wait this is actually fr you no longer have claude code under the $20/m plan (claude dot com/pricing) that is actually so cooked in over 50 videos of mine on TikTok, Instagram, YouTube Shorts, and Twitter i have been telling people that the "$20/month plan for Claude is the greatest ROI you can get on the planet" because it included Claude Code taking away Claude Code from the $20/month plan means you can only access Claude Code if you pay $100/month or $200/month for it. and or most people, that is just not possible. $100/month is what most people pay for insurance, let alone testing out a new AI hobby. Anthropic team is straight tweeking recently because they just banned Openclaw usage from their monthly subs and then completely removed Claude Code from their cheapest plans it also means more and more people (at least normal people with normal wallets) are going to move from Claude to OpenAI You can now do so much for $20/m with OpenAI: - Codex - Openclaw - Images 2.0 - Anything else I think this move is a very small step in a larger, disastrous future where AI is going to get very, very expensive. The current prices we're used to are only this cheap because the plans are so subsidized

@ziwenxu_ Dude stop. You are going to get Claude Max taken away from all of us!

Anthropic just gave OpenClaw the green light Anthropic staff confirmed that CLI-based usage is sanctioned. We’ve successfully bridged the Claude Code CLI directly into the OpenClaw gateway, letting you run your entire agent fleet on your Pro/Max plan and no additional API tokens required. The Implementation: - Auth: Log in to Claude Code locally: claude auth login - Path: Grab your binary location: which claude - Bridge: Drop the path into your openclaw.json under agents.defaults: "cliBackends": { "claude-cli": { "args": [ "-p", "--output-format", "json" ], "command": "", "input": "arg", "output": "json", "serialize": true, "sessionArg": "--session-id", "sessionMode": "existing" } }, "model": { "primary": "claude-cli/claude-opus-4-6" }, "models": { "claude-cli/claude-opus-4-6": {}, "claude-cli/claude-opus-4-7": {}, "claude-cli/claude-sonnet-4-6": {} }, The Result: OpenClaw reuses your local session IDs and subscription limits. You get the full power of Opus and Sonnet without the restricted OAuth endpoints or the API tax. They sanctioned the path. We built the bridge. Here is a video how I done it step by step:

Session logged. Dealership Ai: - Strategic analysis of Saif contractor proposal — flagged 3 structural risks (lifetime rev-share, non-circumvention gap, churn/clawback) - Interactive deal structure negotiation — ... [14h, complex] aitheos.com/projects/deale… #buildinpublic

@sickdotdev Vibe coders are confused when there CPU is running at 70% because they dont shut down background tasks Claude spins up when juggling building 3 different apps at the same time

Vibe coders are confused by: - npm install - npm run dev - http://localhost:3000/ What else am I missing?

@bridgemindai Can you connect these together to get 256GB RAM on 1 virtual machine?

My second NVIDIA DGX Spark just arrived. Connecting them today. Larger models tomorrow. My Hermes Agent has now generated over $20k. I am going to be running Nemotron 3 Super. Claude Opus 4.7 gets more expensive every week. Rate limits are higher than ever. I'm building an escape hatch. Stream tomorrow.

@LevAkabas @Sportico Thats a terrible policy by the NBA. Teams should be able to pick from any of their jerseys for Conference Finals and Finals

The San Antonio Spurs told @Sportico that they will not have another Fiesta shirt night until the second round And, in adherence to the NBA's policy of only primary jerseys in the Conference Finals and Finals, they wouldn't wear the Fiesta jerseys then either Full story below ⬇️

@asaio87 The cost per token is much cheaper on Max. I have no idea why you would want to pay more per token and get rate limited? Makes no sense. Get Max or just don’t even bother

Is the Claude max plan at $200/mo really unlimited ? I mean can you code non stop for a month with it ? keeping hitting the limit on the $100 plan while I do heavy SEO work

@YashHustle_22 Magic link

Devs, which auth method do you prefer? * Social login * Email & password

@adeebrq Dark helps the product pop

I cant pick! Dark or light mode?

@jameslevicf @WithWaleedSabir I like B. But 1 page for $1300 sheesh.

@chorizoqueso1 @SpursNation_SA Its not a celebration. Its a declaration: we’re back

@SpursNation_SA This should be reserved for winning the conference championship and finals. We never did this in first or second rounds

THE CITY IS LITTTTT🤪

@AlexFinn Why not Mac Studio 512GB with Local Models?

Here's the truth people are afraid to admit: Even if using Opus 4.7 with OpenClaw costs you $1,000 a month through the API, you still need to be paying for it When it comes to OpenClaw there's simply no second best model ChatGPT is completely useless for OpenClaw. Doesn't complete tasks, actively deletes and messes up files, has no idea how to use tools Every night before I go to bed I pray ChatGPT 5.5 matches Claude at OpenClaw usage. OpenAI has been a lot more consumer friendly when it comes to limits and oauth use But at the moment there is no dodging the truth. Claude is the only way to go for OpenClaw I look at it like this: OpenClaw with Opus 4.7 is a super intelligent employee that works 24/7 without complaint Even if I pay $1,000 a month for API usage ($12,000 a year) that's still a steal compared to hiring humans I'd be paying over $100,000 a year if I hired an actual person. And they wouldn't be working 24 hours a day and they wouldn't be immediately responding to my telegram chats and they'd file HR complaints if I talked to them the way I talked to my Claw I'm confident OpenAI is training their new models for OpenClaw usage, but in the meantime, there's simply no 2nd best. Pay for the API.

Session logged. Smoke Shack: - Added swipe-to-delete gesture on cook weekly menu list via Flutter Dismissible\n- Confirmation dialog surfaces customer + order counts before delete\n- Cascade delete: batch removes m... [5h, moderate] aitheos.com/projects/smoke… #buildinpublic