What impresses me most about coding agents is that they actually make sense of my fast-typed garble.
0
0
0
14
My workout app noticed I did nothing last week. Now it wants me to open the 0 min report so I can feel the shame twice.
0
0
2
59
For maintainers and on-call teams who had a rough day yesterday (and maybe still today)
Here’s a flower.
Here’s a flower.
0
0
7
205
Raph Terrier retweeted
How disposable software becomes the norm
(originally posted at every.to/thesis-stateme…
Ask ChatGPT to analyze a spreadsheet, and it might write and run a Python script to find the answer. Once you have the answer, you may never need that script again. That is just-in-time, disposable software, and I expect most software run in the cloud will soon look like it.
Today, we build applications around needs we can anticipate. We invest in code review, testing, and deployment pipelines because we expect people to keep using what we ship. When writing a program costs almost nothing and takes seconds, building one to answer a single question starts to make sense.
Source control matters less for programs nobody will revise. We’ll need verification that works as code is generated and executed, without a human reading every line. Permissions and spending limits have to hold even when the program was written moments before it ran, and we’ll still need a record of what it did.
Agents will also take over the infrastructure decisions engineers now make in tools like Terraform. One task needs a few seconds of compute; the next needs a powerful machine for days. The agent will acquire those resources through cloud APIs, within the task’s budget and security constraints, and release them when it’s done. This is application-defined compute: the software chooses its infrastructure as it does the work.
This will make it practical to solve problems that never justified a full software project. A task can have its own software, even if nobody ever needs to do it again.
(originally posted at every.to/thesis-stateme…
Ask ChatGPT to analyze a spreadsheet, and it might write and run a Python script to find the answer. Once you have the answer, you may never need that script again. That is just-in-time, disposable software, and I expect most software run in the cloud will soon look like it.
Today, we build applications around needs we can anticipate. We invest in code review, testing, and deployment pipelines because we expect people to keep using what we ship. When writing a program costs almost nothing and takes seconds, building one to answer a single question starts to make sense.
Source control matters less for programs nobody will revise. We’ll need verification that works as code is generated and executed, without a human reading every line. Permissions and spending limits have to hold even when the program was written moments before it ran, and we’ll still need a record of what it did.
Agents will also take over the infrastructure decisions engineers now make in tools like Terraform. One task needs a few seconds of compute; the next needs a powerful machine for days. The agent will acquire those resources through cloud APIs, within the task’s budget and security constraints, and release them when it’s done. This is application-defined compute: the software chooses its infrastructure as it does the work.
This will make it practical to solve problems that never justified a full software project. A task can have its own software, even if nobody ever needs to do it again.
6
6
35
5,848
Raph Terrier retweeted
Render now supports OIDC for Google Cloud Platform.
Authenticate to GCP via Workload Identity Federation — no static credentials, no manual rotation. Render manages short-lived tokens automatically per service.
To learn more → render.com/docs/oidc
Authenticate to GCP via Workload Identity Federation — no static credentials, no manual rotation. Render manages short-lived tokens automatically per service.
To learn more → render.com/docs/oidc
6
6
24
2,816
Customizing my @work_louder micro v2 with homemade CNC machined brass key caps. The modelling with @OpenAI codex was super quick. Finishing each key by hand takes forever. 8 more to go, then I’ll probably engrave them.
2
0
5
180
Raph Terrier retweeted
Manage your Render infrastructure without leaving the terminal.
Learn how to:
- Trigger deployments
- Manage Postgres & KV stores
- SSH directly into web services
- Install Render Agent Skills
- Automate CI/CD pipelines with non-interactive mode
Full video:
youtube.com/watch?v=pILqLL…
Learn how to:
- Trigger deployments
- Manage Postgres & KV stores
- SSH directly into web services
- Install Render Agent Skills
- Automate CI/CD pipelines with non-interactive mode
Full video:
youtube.com/watch?v=pILqLL…
1
5
23
2,407
Raph Terrier retweeted
You can now build, preview, and deploy code from @cursor_ai Origin repos on Render:
→ Install the Render app from the Origin marketplace
→ Deploy code as a Render service
→ Get automatic builds, PR previews, and zero-downtime deploys
→ Install the Render app from the Origin marketplace
→ Deploy code as a Render service
→ Get automatic builds, PR previews, and zero-downtime deploys
Origin, Cursor’s code hosting platform, now supports over 20 integrations.
Connect your repos to tools for CI, deployments, code review, security, monitoring, and project management.
0
9
24
1,714
Raph Terrier retweeted
What if we treated data pipelines like any other software service?
→ Aggregate first, store second
→ Stand up right-sized services with IaC
→ Leverage production observability
Our latest blog with Turbolytics explores data's DevOps moment: render.com/blog/data-need…
→ Aggregate first, store second
→ Stand up right-sized services with IaC
→ Leverage production observability
Our latest blog with Turbolytics explores data's DevOps moment: render.com/blog/data-need…
4
6
19
1,473
Raph Terrier retweeted
AI agents let you connect LLMs to real-world workflows.
In this course, @ania_kubow shows you how to build and deploy an AI-powered Slackbot.
You'll use Node.js, Express, Slack Bolt, GPT-4, PostgreSQL, and Render to research and get new community members.
U
In this course, @ania_kubow shows you how to build and deploy an AI-powered Slackbot.
You'll use Node.js, Express, Slack Bolt, GPT-4, PostgreSQL, and Render to research and get new community members.
U
13
97
658
33,099
😍
I asked our agent for SaaStr AI Connect live with @HarryStebbings + @rodriscoll today if we wanted to move hosting ... where should we go?
It picked Render over Railway and Vercel. "My pick".
Honestly I had never heard of it. Maybe I should have, but I never had.
Now we'll 's how you get leads in The Agent Age
0
0
2
73
Pretty excited about this new feature. I have a few projects in store that are going to put this to good use. Fill the form if you want early access!
We're launching fast, fully-managed shared filesystems for sandboxes, workflows, and every other service type on @render. Sign up to get early access!
render.com/shared-filesys…
2
0
9
428
Cool conversation @WeAreDevs on the main stage with @anuraggoel and leaders from sentry, blackbox ai, and cognition: Running AI written software in production
0
2
10
571
Come watch @anuraggoel live on stage 1 at @WeAreDevs to learn how application defined compute is simplifying developers life!
0
2
9
365
Come check out the @render booth at @WeAreDevs and chat with the DX team today and tomorrow!
0
1
14
2,317
Come say hi at the booth! The whole DX team will be there to talk shop and show you what we've been building. See you there.
Agent workloads are unpredictable by definition. Is the next request 3 tasks or 400?
Provisioning compute assumes you know the answer.
Our answer is Application-Defined Compute. Render CEO @anuraggoel explains it tomorrow at @WeAreDevs. 11 AM, Stage 1.
Find us at booth #734 after.
0
2
9
210
Raph Terrier retweeted
Agent workloads are unpredictable by definition. Is the next request 3 tasks or 400?
Provisioning compute assumes you know the answer.
Our answer is Application-Defined Compute. Render CEO @anuraggoel explains it tomorrow at @WeAreDevs. 11 AM, Stage 1.
Find us at booth #734 after.
Provisioning compute assumes you know the answer.
Our answer is Application-Defined Compute. Render CEO @anuraggoel explains it tomorrow at @WeAreDevs. 11 AM, Stage 1.
Find us at booth #734 after.
7
8
24
1,607