pratikbin.node

2.7K posts

pratikbin.node banner
pratikbin.node

pratikbin.node

@pratikbin

BareMetalAlchemist @NodeOpsHQ | Contributor @BuildOnNodeOps | slyly outsmarting infra chaos with cloud wizardry | 👨

BLR Katılım Nisan 2014
3.3K Takip Edilen1.6K Takipçiler
pratikbin.node retweetledi
dax
dax@thdxr·
opencode 1.3.0 will no longer autoload the claude max plugin we did our best to convince anthropic to support developer choice but they sent lawyers it's your right to access services however you wish but it is also their right to block whoever they want we can't maintain an official plugin so it's been removed from github and marked deprecated on npm appreciate our partners at openai, github and gitlab who are going the other direction and supporting developer freedom
English
239
450
8.3K
858.4K
pratikbin.node retweetledi
Ben Dicken
Ben Dicken@BenjDicken·
This is why databases are hard. It's also why they are so much fun. Good benchmarking is a particularly powerful tool for discovering performance bottlenecks. I benchmark databases a ton, but this goes for all software where performance matters. (which should be all of it!)
Ben Dicken tweet media
English
4
65
531
12.9K
pratikbin.node
pratikbin.node@pratikbin·
new things this week so far - added image support in messages - private skills from signed s3 zips - upstream error response backprop to api - image retention in chat - pdfs are tricky to handle since support is not there in sdk - testing llmock - e2e CI
pratikbin.node@pratikbin

Lot of new improvements and integrations - apis to CRUD skills - extension - parallel tool calling (v0.58.1) - session timeouts - async persistent storage layer in s3 - final checks for prod rollout Trying to actor based model using rivet.dev

English
0
0
3
48
pratikbin.node
pratikbin.node@pratikbin·
Build it now. Making real-world impact
NodeOps Network@BuildOnNodeOps

Every builder has a unique story to tell. Our fourth episode features a builder with a unique idea: • Project: Riyasat • Builder: Gunjan, Key Account Manager at @amazon • Use case: A platform designed to help people plan destination weddings and inter-caste marriages • Built with CreateOS: Create flow, GitHub deployment • Status: First version is ready and currently being used by the builder What use case should we feature next? #CreateOS #BuildItNow #BuildInPublic

English
1
0
5
171
pratikbin.node
pratikbin.node@pratikbin·
Lot of new improvements and integrations - apis to CRUD skills - extension - parallel tool calling (v0.58.1) - session timeouts - async persistent storage layer in s3 - final checks for prod rollout Trying to actor based model using rivet.dev
pratikbin.node@pratikbin

We are going ahead with pi coder and creating api on top of it, why not agent from scratch? - handles most of the things out of the box - have you create less things to add more things - not fan of TS but its getting things done🥲 still figuring out mcp, extensions....

English
0
0
2
136
pratikbin.node
pratikbin.node@pratikbin·
@badlogicgames Funny, today I was doing research around parallel and programmatic tool calling for our Pi-based web/API project and thought this is an LLM limitation since it requires tool calling serially!! Thank you for dropping this on time
English
0
0
0
27
Mario Zechner
Mario Zechner@badlogicgames·
if an agent emits parallel tool calls, pi used to execute them sequentially. funnily enough, only a handful of people complained. welp, just implemented it. caveat: you'll get flicker if you have many parallel tool calls and their outputs are > terminal height. soon to be released.
English
36
12
246
34.9K
Bhautik 🦒 | NodeOps
Bhautik 🦒 | NodeOps@bhautiktweets·
when your OS writes to disk, it sends: •- sector number •- count of sectors •- the actual data but count is literally derivable from len(data)/512 why? Reason ATA protocol. CPU sent command BEFORE data arrived via DMA. disk needed count upfront to prepare its buffer.
English
1
0
5
33
pratikbin.node retweetledi
dax
dax@thdxr·
i was at a wedding and someone told me they saw how opencode laid off 160 engineers begin is a menace
English
42
8
1.1K
108K
pratikbin.node
pratikbin.node@pratikbin·
You can't beat me🥲 ps: please dont
pratikbin.node tweet media
English
1
0
4
238
tobi lutke
tobi lutke@tobi·
OK, well. I ran /autoresearch on the the liquid codebase. 53% faster combined parse+render time, 61% fewer object allocations. This is probably somewhat overfit, but there are absolutely amazing ideas in this.
tobi lutke tweet media
English
102
171
2.8K
938.5K
CopilotKit🪁
CopilotKit🪁@CopilotKit·
✨ Introducing LLMock: A deterministic mock LLM server for testing Test your AI powered apps reliably, without burning money on real API calls or fighting non-deterministic outputs in CI. Open-sourced for the community. llmock.copilotkit.dev
English
16
62
464
77.8K
Steve Faulkner
Steve Faulkner@southpolesteve·
OpenCode hasn't hyped it up but remote server support is really good. You can see where this is going. Excited to see the end state of what they build.
English
16
5
326
79.7K
Bhautik 🦒 | NodeOps
Bhautik 🦒 | NodeOps@bhautiktweets·
After learning executor stuff, I want every website should have MCP, from music playing to monitoring🤣
Bhautik 🦒 | NodeOps tweet media
English
1
0
0
27
Karan Sharma
Karan Sharma@mrkaran_·
and if you need an OSS variant of the same thing - i've built hodor. Uses the excellent @badlogicgames 's pi-sdk, supports gitlab/github, and you can use any model (even bedrock with custom profiles to track costs)! You can also load custom "skills" per repo/CI job - enforce your own conventions, security rules, or review checklists. This is what we @zerodha have been using for the past few weeks and it has caught some really non-obvious subtle edge cases/bugs. github.com/mr-karan/hodor
Claude@claudeai

Agents search for bugs in parallel, verify each bug to reduce false positives, and rank bugs by severity. You get one high-signal summary comment plus inline flags.

English
12
12
157
21.1K
Naval
Naval@naval·
A “computer” used to be a job title. Then a computer became a thing humans used. Now a computer is becoming a thing computers use.
English
1.3K
2.9K
29.8K
38.8M