tony @Cephalization

fullstack software engineer https://t.co/gjCwES66fj https://t.co/RJ7L6TD8Hb anthonypowell.me Joined April 2014

Tweets

2K
Followers

179
Following

1K
Likes

20K

tony @Cephalization

5 days ago

@threepointone @mattzcarey perfect fit, very cool role

0 0 1 60 0

View Details

tony @Cephalization

5 days ago

@kwuchu sorry man 😔

0 0 1 58 0

View Details

tony @Cephalization

a week ago

@rachelnabors I should set one of those up on my mac mini, cool

1 0 1 20 0

View Details

Is this essentially pokemon training for AI engineers? Two people have the same pikachu, but depending on how they train it, they have different strengths and specialties? Can I trade my web development prime agent with a system eng prime agent and teach it new things?

Prime Intellect @PrimeIntellect

2 weeks ago

Introducing Prime Agent: A self-improving RLM harness for coding and long-running autonomous tasks. Designed to be both token-efficient and expressive through programmatic tool calling, context as a variable, multi-agent messaging, and a self-modifiable harness state.

399 833 8K 3.0M 7K

0 0 2 143 0

View Details

tony @Cephalization

2 weeks ago

@stevekrouse 🫡

0 0 0 56 0

View Details

tony @Cephalization

2 weeks ago

@KentonVarda cloudfLairs

0 0 13 430 0

View Details

tony @Cephalization

2 weeks ago

@adamwathan Makes me think more and more that @AmpCode made the right bet by abstracting away the models under the hood

0 0 0 235 0

View Details

Ryan Dahl @rough__sea

2 weeks ago

Introducing celld: a self-hosted, distributed Durable Objects and Workers implementation - celld = V8 + S3 + SQLite + LTX+ Tokio - programmed with exactly the Cloudflare Workers/DO JavaScript APIs and config - order of magnitude cheaper at scale - writes are durable before they're acknowledged, RPO=0 - only depends on an S3 compatible bucket - no other infrastructure - ~4 MB memory per cell instance overhead - hibernated cells cost ~nothing, just their S3 storage - open source, Apache 2 licensed

116 308 3K 274K 2K

View Details

tony @Cephalization

2 weeks ago

@rough__sea @boristane oh this is incredible, I have been waiting for something like this for years

0 0 3 2K 0

View Details

tony @Cephalization

3 weeks ago

@wesbos I notice it the most when an app opens finder as a file picker. Ableton 12 opening a new project from the file menu for example is intolerable. Opening my docker folder and attempting to double click folders open as well

0 0 1 244 0

View Details

tony @Cephalization

3 weeks ago

@cassidoo yes, they are great

0 0 1 106 0

View Details

tony @Cephalization

3 weeks ago

@joshuatopolsky @Rivian R2 or R1S? I recently got an R1S and love it

1 0 0 954 0

View Details

tony @Cephalization

4 weeks ago

@mikeldking @mitchellh im stuck on how nice the tab icons look

0 0 1 36 0

View Details

sunil pai @threepointone

4 weeks ago

new post: one document, two hands sunilpai.dev/posts/one-docu… wherein I propose an alternative interaction model for agentic apps, embedding so called "coding agents" harnesses into regular apps. (a written version of my talk at localfirstconf)

19 23 256 58K 285

View Details

Mikyo @mikeldking

a month ago

Over the past 6 months we've maniacally prepped our repos to be coding agent friendly. Here are some things that worked. 1. Make CI blazing fast. Use every Rust, Zig, or Go ported tool that lets agents verify their work. This means UV, oxfmt, Typescript 7. Move integration tests to post merge hooks. 2. Trigger coding agents automatically based on triage labels. A coding agent should setup a proof of concept or repro steps automatically so an engineer can pick up the issue seamlessly. 3. Setup crons for things devs hate doing. Setup agents to fill SDK gaps, skill tuning, filling in critical regression checks. 4. Give agents ways to prove their work. Add screenshotting skills, agent-browswer, cloud storage for storing assets. 5. Make it possible to hermetically deploy your app, preferably multiple at a time. If a coding agent can deploy the app locally, the faster it can work. 6. Give the agents realistic production "simulation" data. Agents will work much better when they are working against data that looks like how your users use the product.