Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

Replicas is a background coding agent that lives in sandboxed VMs and works like a teammate: it runs codebases locally with all your dependencies and tooling, iterates until it's satisfied, and opens a PR. Delegate to it from Slack, Linear, or GitHub. Congrats on the launch, Connor Loi!

ycombinator's profile picture

Y Combinator

1,653,446 subscribers

22,356 Aufrufe • vor 7 Monaten •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

32 Kommentare

Profilbild von Nick Khami
Nick Khamivor 7 Monaten

absolutely loving replicas here at Mintlify!

Profilbild von Saai Arora
Saai Aroravor 7 Monaten

the uw hoodie is aura

Profilbild von Squeak
Squeakvor 7 Monaten

woah this is sick, congrats on the launch!

Profilbild von angela
angelavor 7 Monaten

CONGRATS CONNOR @connortbot 🐐🚀

Profilbild von pooria
pooriavor 7 Monaten

Q: Where are you hosting your sandboxed VMs and do you use Docker images or keep warm sandboxes? @connortbot (can dm too)

Profilbild von pranav
pranavvor 7 Monaten

goated product

Profilbild von Alisa
Alisavor 7 Monaten

congrats @connortbot on the launch, highly recommend trying replicas

Profilbild von fuelvin
fuelvinvor 7 Monaten

Awesome stuff!

Profilbild von Stephen Ni
Stephen Nivor 7 Monaten

CONGRATS ON THE LAUNCH! @connortbot

Profilbild von Adi Singh
Adi Singhvor 7 Monaten

love this

Profilbild von cackles (jeff weisbein)
cackles (jeff weisbein)vor 7 Monaten

very cool! love to see more people building this way. been running a similar setup with openclaw — agents that live on your own machines, delegate over ssh, and handle the full dev loop. the key difference is keeping it on hardware you control vs cloud sandboxes. open source too

Profilbild von SOUHAIL
SOUHAILvor 7 Monaten

Congrats @connortbot

Profilbild von ben
benvor 7 Monaten

let's go Connor 📈

Profilbild von UNI Network Group
UNI Network Groupvor 7 Monaten

Game-changing for dev teams—your AI teammate that codes, iterates, and delivers without missing a beat! 💻🚀

Profilbild von audacity
audacityvor 7 Monaten

how is this at all useful

Profilbild von Akira
Akiravor 7 Monaten

background agents r the future 🗣️

Profilbild von Ruhan Ponnada
Ruhan Ponnadavor 7 Monaten

LFGGG

Profilbild von shams haroon
shams haroonvor 7 Monaten

@connortbot congrats on the launch 🤝

Profilbild von Kush
Kushvor 7 Monaten

@connortbot Amazing @connortbot

Profilbild von Marchen
Marchenvor 7 Monaten

the sandboxed VM approach is smart. running in the actual dev environment with real deps solves a big failure mode that pure API-based agents hit constantly

Profilbild von Shankar Radhakrishnan
Shankar Radhakrishnanvor 7 Monaten

Congrats @connortbot!!

Profilbild von Damir Wallener
Damir Wallenervor 7 Monaten

Yet another app in the process of being swallowed by Codex/Claude…

Profilbild von Carlton Aikins
Carlton Aikinsvor 7 Monaten

@connortbot is the goat

Profilbild von matija
matijavor 7 Monaten

How is it different from Cursor background agents

Profilbild von Jonathan Wang
Jonathan Wangvor 7 Monaten

lighting was a paid actor

Profilbild von Prince Ruhul
Prince Ruhulvor 7 Monaten

See you soon with Prevalid.

Profilbild von dnu
dnuvor 7 Monaten

worktree isolation is the easy part. cold start and secret scoping are where background agents usually get ugly once npm installs and CI creds enter the loop. are these VMs fully fresh per task, or do you keep a warm cache somewhere?

Profilbild von David Goodfellow
David Goodfellowvor 7 Monaten

@connorbot Does this work with multi repo set ups? with isolated worktrees at the multi repo level?

Profilbild von Matt
Mattvor 7 Monaten

Congrats on the Launch!

Profilbild von leyland
leylandvor 7 Monaten

lets gooo, love replicas

Profilbild von nikolas
nikolasvor 7 Monaten

LFG

Profilbild von LightShift Studio
LightShift Studiovor 7 Monaten

Hey there!

Ähnliche Videos

Introducing Replicas V1, the background coding agent built for you. Replicas is a background coding agent that: - Lives in its own sandbox - Runs your codebase locally - Accessible from anywhere (PR reviews, Linear, Slack, API) Replicas
1:29

Introducing Replicas V1, the background coding agent built for you. Replicas is a background coding agent that: - Lives in its own sandbox - Runs your codebase locally - Accessible from anywhere (PR reviews, Linear, Slack, API) Replicas

Connor Loi

16,928 Aufrufe • vor 7 Monaten

Introducing Replicas V3: the most flexible and cost-efficient cloud agent on the market. - Run Claude Code, Codex inside VMs - Bring any subscription or API keys - Delegate from Slack, Linear, desktop app, or your phone Now proactive, and more powerful than ever.
2:02

Introducing Replicas V3: the most flexible and cost-efficient cloud agent on the market. - Run Claude Code, Codex inside VMs - Bring any subscription or API keys - Delegate from Slack, Linear, desktop app, or your phone Now proactive, and more powerful than ever.

Connor Loi

42,558 Aufrufe • vor 1 Monat

Introducing Shadow - a background coding agent that runs on VMs allowing you to run code from anywhere asynchronously. Try it now at
1:00

Introducing Shadow - a background coding agent that runs on VMs allowing you to run code from anywhere asynchronously. Try it now at

Elijah Kurien

50,680 Aufrufe • vor 1 Jahr

Now ready for deployment: Cursor Delegate any Linear issue to Cursor, and a background agent will get to work. Once the task is complete, the Cursor agent will update the issue automatically with a PR.
0:42

Now ready for deployment: Cursor Delegate any Linear issue to Cursor, and a background agent will get to work. Once the task is complete, the Cursor agent will update the issue automatically with a PR.

Linear

2,515,115 Aufrufe • vor 1 Jahr

from CambioML is your AI teammate that works side-by-side with you on a secure desktop to automate data-heavy, multi-step workflows, without any integrations or coding. Congrats on the launch, Rachel Hu and Lingjie (Kimi) Kong!
2:39

from CambioML is your AI teammate that works side-by-side with you on a secure desktop to automate data-heavy, multi-step workflows, without any integrations or coding. Congrats on the launch, Rachel Hu and Lingjie (Kimi) Kong!

Y Combinator

20,654 Aufrufe • vor 1 Jahr

Factory 1.8 is here. And with it come new ways to automate and delegate software development with your Droids. Now in your favorite tools like Linear, Slack, and Sentry.
2:56

Factory 1.8 is here. And with it come new ways to automate and delegate software development with your Droids. Now in your favorite tools like Linear, Slack, and Sentry.

Factory

47,076 Aufrufe • vor 1 Jahr

Are you a 100x engineer but a 1x tester with AI? Arga Labs closes that gap. Arga spins up a per-PR sandbox with twins (Stripe, Slack, etc.) and in-memory dependencies (DB, Redis). Run auto-generated E2E tests in the sandbox and send failures back to your agent so it fixes bugs before you see them. Congrats on the launch, Phillip Li & Akira!
1:18

Are you a 100x engineer but a 1x tester with AI? Arga Labs closes that gap. Arga spins up a per-PR sandbox with twins (Stripe, Slack, etc.) and in-memory dependencies (DB, Redis). Run auto-generated E2E tests in the sandbox and send failures back to your agent so it fixes bugs before you see them. Congrats on the launch, Phillip Li & Akira!

Y Combinator

50,252 Aufrufe • vor 6 Monaten

🚨 The END of IDE era. Meet Nova v1 we're calling it Flash - your AI engineering teammate. You hand off the work, watch it happen live, and decide what ships. - Threads for your codebase work, to chase bugs, and delegate changes - Automations that run on a schedule, a webhook, or GitHub, Slack, and Linear events - PRs inside Nova with diffs and Supercode Review - AI review that flags bugs and security issues before you merge - GitHub, Slack, and Linear connected through Composio - Cloud machines with concurrent RAM and upto 64gb - Codex, Claude, connection right into Nova and your choice of models - A live cloud browser you can take over and hand back All in one tool. Try it today 👇
2:59

🚨 The END of IDE era. Meet Nova v1 we're calling it Flash - your AI engineering teammate. You hand off the work, watch it happen live, and decide what ships. - Threads for your codebase work, to chase bugs, and delegate changes - Automations that run on a schedule, a webhook, or GitHub, Slack, and Linear events - PRs inside Nova with diffs and Supercode Review - AI review that flags bugs and security issues before you merge - GitHub, Slack, and Linear connected through Composio - Cloud machines with concurrent RAM and upto 64gb - Codex, Claude, connection right into Nova and your choice of models - A live cloud browser you can take over and hand back All in one tool. Try it today 👇

Yash

10,339 Aufrufe • vor 6 Tagen

A bug enters triage. Linear instantly labels it, runs a full code investigation, and opens a PR for your team to review. Automated triage that actually closes the loop. 🔁
0:54

A bug enters triage. Linear instantly labels it, runs a full code investigation, and opens a PR for your team to review. Automated triage that actually closes the loop. 🔁

Linear

62,161 Aufrufe • vor 4 Monaten

Just created a PR from a github issue via a serverless background agent, and it's entirely open-source at humanlayer we have been thinking a lot about what good open-source software factory building blocks look like to push on this I built something I've wanted for a long time: a hands-outside-the-box approach of running a lightweight agent in a runtime with a virtual fs, and only using a container when necessary - faster, cheaper, etc combined with really good webhook / event bindings for GitHub, slack, and linear we built two components for it: 1. - effect-native webhook subscriptions for slack/linear/github; like the Vercel chat SDK but effect + more opinionated about software factories; handles threading & debouncing event deliveries and stuff like that 2. mentions & other events spin up our Fold harness ( think kinda like Pi in effect; fully isomorphic, runs anywhere) infra is built on Cloudflare workers and durable objects via alchemy - each thread/channel/issue/pr gets its own mailbox, subject to delivery policies (queue, debounce, batch) mailbox messages wake a fold agent session running in a Durable object, without a sandbox to start it uses cloudflare/computer for a virtual DO-backed filesystem for each issue/PR's git branch the agent can run most commands it wants against the DO's virtual filesystem and the agent can boot a container with the FS mounted into it via FUSE only when it decides it needs one to run heavier commands tools are generated from the webhook event so the agent gets tightly-scoped tools to e.g. create a pr from the issue it's on, or to list failing checks only from the PR it's bound to; plus it's easy to interleave deterministic code it has live progress updates, can react to issues in its thread, etc (linear + slack modules support their respective streaming agent activity & agent plan APIs; for github we fudge it a bit by editing issues in live time) it even has an API so agent work can be handed off to another system (e.g. a k8s cluster that has super thick sandboxes if you need > standard-4 sized-containers from cloudfalre) and that system can post back streaming updates & update the status of the agent run, so it's pretty de-coupled it's wired to automatically investigate CI failures and some other neat stuff open-source at a good example of it in action:
3:44

Just created a PR from a github issue via a serverless background agent, and it's entirely open-source at humanlayer we have been thinking a lot about what good open-source software factory building blocks look like to push on this I built something I've wanted for a long time: a hands-outside-the-box approach of running a lightweight agent in a runtime with a virtual fs, and only using a container when necessary - faster, cheaper, etc combined with really good webhook / event bindings for GitHub, slack, and linear we built two components for it: 1. - effect-native webhook subscriptions for slack/linear/github; like the Vercel chat SDK but effect + more opinionated about software factories; handles threading & debouncing event deliveries and stuff like that 2. mentions & other events spin up our Fold harness ( think kinda like Pi in effect; fully isomorphic, runs anywhere) infra is built on Cloudflare workers and durable objects via alchemy - each thread/channel/issue/pr gets its own mailbox, subject to delivery policies (queue, debounce, batch) mailbox messages wake a fold agent session running in a Durable object, without a sandbox to start it uses cloudflare/computer for a virtual DO-backed filesystem for each issue/PR's git branch the agent can run most commands it wants against the DO's virtual filesystem and the agent can boot a container with the FS mounted into it via FUSE only when it decides it needs one to run heavier commands tools are generated from the webhook event so the agent gets tightly-scoped tools to e.g. create a pr from the issue it's on, or to list failing checks only from the PR it's bound to; plus it's easy to interleave deterministic code it has live progress updates, can react to issues in its thread, etc (linear + slack modules support their respective streaming agent activity & agent plan APIs; for github we fudge it a bit by editing issues in live time) it even has an API so agent work can be handed off to another system (e.g. a k8s cluster that has super thick sandboxes if you need > standard-4 sized-containers from cloudfalre) and that system can post back streaming updates & update the status of the agent run, so it's pretty de-coupled it's wired to automatically investigate CI failures and some other neat stuff open-source at a good example of it in action:

Kyle Mistele 🏴‍☠️

32,746 Aufrufe • vor 27 Tagen

.EigenPal is AGI for backoffice paperwork: describe a process to automate, give it test data, and the AI agent iterates until the workflow is perfect. Try on Congrats on the launch Jedr Blaszyk & Matej Novak!
1:39

.EigenPal is AGI for backoffice paperwork: describe a process to automate, give it test data, and the AI agent iterates until the workflow is perfect. Try on Congrats on the launch Jedr Blaszyk & Matej Novak!

Y Combinator

21,885 Aufrufe • vor 9 Monaten

we built pookie it's a helper that lives inside Slack it can search messages, generate memes, run code, and connect to your tools (Linear, GitHub, Stripe, even my bank!) fully open source and self hostable
1:05

we built pookie it's a helper that lives inside Slack it can search messages, generate memes, run code, and connect to your tools (Linear, GitHub, Stripe, even my bank!) fully open source and self hostable

Aiden Bai

45,831 Aufrufe • vor 6 Monaten

The new GitHub Copilot coding agent can work in the background while you focus on other tasks. Assign it tasks via GitHub Issues or ad hoc in Visual Studio Code. Watch this demo and then optimize your workflow with the Copilot coding agent. 🚀 ⬇️
2:03

The new GitHub Copilot coding agent can work in the background while you focus on other tasks. Assign it tasks via GitHub Issues or ad hoc in Visual Studio Code. Watch this demo and then optimize your workflow with the Copilot coding agent. 🚀 ⬇️

GitHub

55,411 Aufrufe • vor 1 Jahr

You can now delegate tasks to Cursor directly from Notion. It's built on the Cursor SDK, so every cloud agent runs on the same models, harness, and runtime that power Cursor. @​Cursor on any spec or assign it a task to open a PR your whole team can review.
0:37

You can now delegate tasks to Cursor directly from Notion. It's built on the Cursor SDK, so every cloud agent runs on the same models, harness, and runtime that power Cursor. @​Cursor on any spec or assign it a task to open a PR your whole team can review.

Cursor

324,657 Aufrufe • vor 4 Monaten

Today we're launching Amplitude Custom Agents. Teammates that automate recurring workflows on your product data, 24/7. The signal your team needs already exists in Amplitude. Getting it to the right person, in the right format, before the moment passes still requires someone to do the assembly. Custom Agents change that: - Describe the job once: give the agent a name, instructions, and output format and it runs that workflow on schedule - Connects to the tools your team already works in: Slack, Jira, Linear, GitHub, and more - Runs on behavioral data from Amplitude, so the output reflects what your users actually did - Pushes it to systems where you work: a filed ticket in Linear, a Slack message, a doc ready before a customer call If you're lacking inspiration, here are a bunch of use cases you can start from: Have fun!
0:35

Today we're launching Amplitude Custom Agents. Teammates that automate recurring workflows on your product data, 24/7. The signal your team needs already exists in Amplitude. Getting it to the right person, in the right format, before the moment passes still requires someone to do the assembly. Custom Agents change that: - Describe the job once: give the agent a name, instructions, and output format and it runs that workflow on schedule - Connects to the tools your team already works in: Slack, Jira, Linear, GitHub, and more - Runs on behavioral data from Amplitude, so the output reflects what your users actually did - Pushes it to systems where you work: a filed ticket in Linear, a Slack message, a doc ready before a customer call If you're lacking inspiration, here are a bunch of use cases you can start from: Have fun!

Spenser Skates

11,426 Aufrufe • vor 4 Monaten

Introducing coding sessions. Linear Agent can now triage issues, investigate the cause, write the fix, open a PR, and bring the code back for review. All shared with your team in Linear.
7:06

Introducing coding sessions. Linear Agent can now triage issues, investigate the cause, write the fix, open a PR, and bring the code back for review. All shared with your team in Linear.

Linear

309,083 Aufrufe • vor 4 Monaten

How is Cognition working with Devin in Slack Code? 👀👇 Devin follows the work as it happens, picks up a task from the engineering channel, opens a PR with a demo video, and brings in a designer to finish the UI. All without leaving Slack or waiting to be tagged. The agent isn’t off doing its own thing. The whole team is right there, building with it. That's the beauty of multiplayer AI in Slack.
2:19

How is Cognition working with Devin in Slack Code? 👀👇 Devin follows the work as it happens, picks up a task from the engineering channel, opens a PR with a demo video, and brings in a designer to finish the UI. All without leaving Slack or waiting to be tagged. The agent isn’t off doing its own thing. The whole team is right there, building with it. That's the beauty of multiplayer AI in Slack.

Slack

51,508 Aufrufe • vor 2 Monaten

🇺🇸 xAI launched Grok Build, a coding agent that runs directly from your terminal, now in early beta for SuperGrok and X Premium Plus subscribers. It lets you plan and approve changes before anything runs, works with tools you already use like MCP servers and plugins, and can spin up multiple subagents working in parallel on larger tasks. Some users say it's already competing with Claude Code and GitHub Copilot Workspace. Source: xAI
0:43

🇺🇸 xAI launched Grok Build, a coding agent that runs directly from your terminal, now in early beta for SuperGrok and X Premium Plus subscribers. It lets you plan and approve changes before anything runs, works with tools you already use like MCP servers and plugins, and can spin up multiple subagents working in parallel on larger tasks. Some users say it's already competing with Claude Code and GitHub Copilot Workspace. Source: xAI

Mario Nawfal

99,324 Aufrufe • vor 5 Monaten

Alibaba just dropped an AI that doesn’t just write code. It runs it. Tests it. Fixes it. Until it actually works. Here’s what makes Qwen 3 Coder Next different: → Executes code, not just autocomplete → Runs tests automatically → Iterates until all tests pass → Handles multi-file repositories → Works fully offline This is a real coding agent, not a text generator. Save this video, you’ll rethink how coding gets done. Want the SOP? DM me. 💬
8:15

Alibaba just dropped an AI that doesn’t just write code. It runs it. Tests it. Fixes it. Until it actually works. Here’s what makes Qwen 3 Coder Next different: → Executes code, not just autocomplete → Runs tests automatically → Iterates until all tests pass → Handles multi-file repositories → Works fully offline This is a real coding agent, not a text generator. Save this video, you’ll rethink how coding gets done. Want the SOP? DM me. 💬

Julian Goldie SEO

20,956 Aufrufe • vor 8 Monaten