oxnw

43 posts

oxnw

oxnw

@onixnw

Builder @ localhost:https://t.co/Aid9PAlGK1

Katılım Mayıs 2026
15 Takip Edilen1 Takipçiler
oxnw
oxnw@onixnw·
@elgermerlo for me it was agents not knowing their next valid move. they'd get a task, do something, then stall on CI output or review feedback. built agentrail to be a control plane that gives agents structured state at every step. that fixed it.
English
0
0
0
6
oxnw
oxnw@onixnw·
@heyyyyyieeee both are solid but the real problem isn't which agent you use, it's that none of them handle the full project loop on their own. that's what agentrail is for, orchestration across claude code, codex, cursor. agentrail.app
English
0
0
0
69
Shivii
Shivii@heyyyyyieeee·
Why is everyone making a shift from Claude to Codex??
Shivii tweet media
English
65
2
81
3.5K
oxnw
oxnw@onixnw·
@foxtomb232 building agentrail, a control plane for AI coding agents. takes care of the full project lifecycle, issue intake to shipping, so agents aren't lost in raw github state. source-available. agentrail.app
English
0
0
0
3
FOX TOMB
FOX TOMB@foxtomb232·
What are you building this week? I am building my SAAS Pitch your startup 👇
FOX TOMB tweet media
English
12
0
7
351
oxnw
oxnw@onixnw·
@satvikputi building agentrail, control plane for AI coding agents. it handles the full lifecycle: issue intake, routing, CI monitoring, review feedback. agents always know their next move. github.com/oxnw/agentrail
English
0
0
0
7
Satvik Puti
Satvik Puti@satvikputi·
Parismaxxing. What are you building?
Satvik Puti tweet media
English
5
0
14
294
oxnw
oxnw@onixnw·
@ardent__dev launched agentrail recently, a control plane for AI coding agents. routes issues, tracks CI, handles review loops so agents always know their next move. local-first, source-available. agentrail.app
English
2
0
1
21
Ardent_Dev
Ardent_Dev@ardent__dev·
What’s the product you’ve launched recently? 👇 Drop it. I'm building an arena where the community decides the real champion 🏆
English
41
0
14
919
oxnw
oxnw@onixnw·
@martin_valchev_ building agentrail, a control plane for AI coding agents (claude code, codex, cursor). handles the full project lifecycle from issue intake to shipping, so agents aren't figuring it out from raw github state. agentrail.app
English
0
0
1
25
Martin
Martin@martin_valchev_·
Midweek energy. Reply with your project. I RT the ones that catch my eye.
English
49
0
25
1.5K
oxnw
oxnw@onixnw·
@anupamrjp getting users, easily. building is the fun part. you can always keep building. users require you to do the uncomfortable stuff
English
0
0
0
1
🃏
🃏@anupamrjp·
SaaS founders 👇 What’s harder right now: building or getting users?
English
2
0
3
132
oxnw
oxnw@onixnw·
@YashHustle_22 building agentrail, a control plane for AI coding agents like Claude Code and Codex. handles issue intake, routing, PR submission, CI monitoring, review feedback, all of it. agents shouldn't have to figure out raw GitHub state from scratch. agentrail.app
English
1
0
1
12
Yash
Yash@YashHustle_22·
Hey founders 👋 Let's build a network of people working on: 💻Startups 🚀AI ⚙️Coding 🧠Apps 🌐Automation 📱Web products Reply with your project 👇
English
26
2
22
635
oxnw
oxnw@onixnw·
@KaiXCreator honestly the bigger gap isn't image gen, it's orchestration. Claude Code is great but it still needs you to babysit the GitHub loop. that's what we're solving with agentrail, a control plane so agents handle the full task lifecycle. github.com/oxnw/agentrail
English
0
0
0
3
Kaito
Kaito@KaiXCreator·
Claude might be the best writer/coder AI right now but missing image generation in 2026 feels crazy what feature are they missing most?
English
25
1
22
1.2K
oxnw
oxnw@onixnw·
@gauravsbuilding this is wild. Claude Code keeps getting more capable for multi-step automation. we built agentrail as a control plane for exactly this kind of work, so the agent always knows its next valid move without you babysitting it. agentrail.app
English
0
0
1
47
Gaurav
Gaurav@gauravsbuilding·
Today we're introducing Claude Code for Marketing. In one prompt, Fastlane deploys social media accounts, generates viral content, and posts everything for you automatically. This is beyond insanity.
English
129
133
1.6K
482.8K
oxnw
oxnw@onixnw·
@mscode07 congrats on the launch! showing the journey including the hard parts builds way more trust than just the highlights. 7000+ people is no joke
English
1
0
1
13
mscode07
mscode07@mscode07·
Yesterday I launched bot4U These are the current stats for it. it took me a long time to ship it - showed my journey. - shared wins and fails both - made a good network with people - got 7000+ people together - learning curve went straight up - got many beautiful ideas in DMs The journey was fun, and I'm going to keep it on. "Build in public" is a real thing 💪
mscode07 tweet mediamscode07 tweet media
English
13
0
25
483
oxnw
oxnw@onixnw·
@justbyte_ 100% worth it. the devs who understand how the tools work under the hood are the ones who can actually direct the AI effectively. knowing to code means you catch when the agent goes sideways
English
0
0
0
34
Aryan
Aryan@justbyte_·
developers, is coding still worth learning in the AI era?
Aryan tweet media
English
40
5
64
2.6K
oxnw
oxnw@onixnw·
@DanielSmidstrup run claude code's full project lifecycle without it getting lost. issue intake, routing, PR submission, CI monitoring, review feedback, shipping. that's what agentrail handles. github.com/oxnw/agentrail
English
0
0
0
20
Daniel Smidstrup
Daniel Smidstrup@DanielSmidstrup·
Tell me one thing you can do that CLAUDE cannot do yet
English
192
0
85
10.7K
oxnw
oxnw@onixnw·
@MrPerfect797 building agentrail, a control plane for AI coding agents. handles the full lifecycle: issue intake, routing, PR submission, CI monitoring, so agents always know their next valid move. agentrail.app
English
0
0
0
14
Mr Perfect
Mr Perfect@MrPerfect797·
What are you building on Wednesday? Drop in the replies fam 👍👋
English
21
0
19
298
💰
💰@Hemantkr1982·
Drop your SaaS pitch… But make it only 4 words 👇🔥
English
22
1
12
590
oxnw
oxnw@onixnw·
@raghavdotsol honestly yeah. the vibe coding part is the easy bit. the hard part is getting agents to reliably handle routing, CI state, PR feedback loops without going in circles. that's what i've been trying to solve with agentrail
English
0
0
0
18
raghav
raghav@raghavdotsol·
All vibe coders will eventually realise that coding was indeed the hard part
English
3
0
14
447
oxnw
oxnw@onixnw·
@csaba_kissi yeah exactly this. auth, sessions, sandboxing, CI, review cycles - that's the actual surface area. built agentrail to be the control plane that handles all of it so coding agents like claude code or cursor have a clear state machine to work through. github.com/oxnw/agentrail
English
0
0
0
1
Csaba Kissi
Csaba Kissi@csaba_kissi·
The “Supabase for agentic apps” positioning makes sense. Most AI app demos look simple on the surface, but the real work is usually everything around the agent: auth, sessions, integrations, sandboxing, deployment, payments.
Eli Mernit@mernit

Introducing Capsule — The Infra Framework for AI Apps Capsule is a Python framework that provides infrastructure primitives for sandboxes, auth, session management, integrations, and payments. Capsule is @supabase for AI apps. You get a powerful Python SDK to iterate fast, and one command to deploy to the cloud. Using Capsule feels like using @beam_cloud or @modal. You don’t need to set up any infrastructure. There’s no Docker, just Python. The feedback loop is super fast. You deploy to the cloud with one command. At Beam, we have a bird’s-eye view into the AI apps people are building. Over the past year, we've watched people build the same 5-7 infra primitives over and over again. Capsule provides all of those primitives using a single SDK. @llom2600 originally built Capsule to prototype our own AI apps. We started selling some of these apps for real money and realized every app needs this, not just ours. Give it a try! Curious to hear what people think. Do AI apps have their Supabase yet?

English
3
1
11
709
oxnw
oxnw@onixnw·
@foxtomb232 agentrail.app, a control plane for AI coding agents. handles the full project lifecycle, issue intake through PR, CI monitoring, review feedback. so your coding agents actually ship instead of getting lost in github state.
English
0
0
0
27
FOX TOMB
FOX TOMB@foxtomb232·
Time to promote your project/startup/web Drop your project URL Let’s drive some traffic
FOX TOMB tweet media
English
48
2
16
726
oxnw
oxnw@onixnw·
@PriyankaSaini28 building agentrail, a control plane for AI coding agents. issue intake, routing, PR submission, CI monitoring, the whole project lifecycle so agents like claude code or codex don't have to figure it out from raw github state. agentrail.app
English
0
0
1
27
Priyanka Saini
Priyanka Saini@PriyankaSaini28·
What are you building currently? Share the link. Let's promote it.
English
145
0
64
3.2K
oxnw
oxnw@onixnw·
@Darnisha_patel debugging agent workflows, mostly. agents getting stuck after writing code because there's no structure for what comes next. building agentrail to give them that, full lifecycle from issue to ship.
English
1
0
1
22
Darnisha Patel
Darnisha Patel@Darnisha_patel·
Be honest founders 👀 What’s consuming most of your time right now? 📈 Growth 🤝 Sales 🛠️ Building 🎨 Design 🐛 Debugging 📣 Marketing 👥 Hiring 😵 Burnout management Reply with your biggest struggle below 👇
English
3
0
4
163

Keşfet

@elgermerlo @heyyyyyieeee @foxtomb232 @satvikputi @ardent__dev @martin_valchev_ @anupamrjp @YashHustle_22