You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Make sure to call `browser.close()` at the end of your action to ensure the browser connection is properly closed.
136
132
</Info>
137
133
138
-
## Browser frameworks: environment variables
139
-
140
-
Browser frameworks sometimes require environment variables (e.g. LLM API keys). Set them when [deploying](/launch/deploy#environment-variables) your app.
141
-
142
-
## Next steps
134
+
## Web agent frameworks: environment variables
143
135
144
-
Once you've created your browser automation app:
145
-
1.[Deploy](/launch/deploy) it on Kernel's cloud platform
146
-
2. Learn how to [invoke](/launch/invoke) your app's actions
147
-
3. Explore [browser configuration options](/launch/browser-persistence) for advanced use cases
136
+
Web agent frameworks sometimes require environment variables (e.g. LLM API keys). Set them when [deploying](/launch/deploy#environment-variables) your app.
Copy file name to clipboardExpand all lines: build/remote-browsers.mdx
+1-1Lines changed: 1 addition & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -2,7 +2,7 @@
2
2
title: "Remote Browsers"
3
3
---
4
4
5
-
Kernel [browsers](/info/unikernels) were designed to be lightweight, fast, and efficient for cloud-based browser automation at scale. They can be used as part of the Kernel [app platform](/build/develop) or connected a **remote service** with Chrome DevTools Protocol.
5
+
Kernel browsers were designed to be lightweight, fast, and efficient for cloud-based browser automation at scale. They can be used as part of the Kernel [app platform](/build/develop) or connected to a **remote service** with Chrome DevTools Protocol.
6
6
7
7
Remote Kernel browsers have the same capabilities as on our app platform, including:
- 💻 **Skills: Bun / Typescript, Go, Stainless, MySQL (PlanetScale)**
9
9
10
10
## About Kernel
11
11
12
-
We're building *the* developer platform that lets AI agents use real applications, starting with browsers. Our edge is an infrastructure platform that's extensible, observable, and built for scale from day one. We've shipped foundational tech, landed paying customers, and are well-funded while still in stealth.
12
+
We're building *the* developer platform that lets AI agents use applications, starting with browsers. Our edge is an infrastructure platform that's extensible, observable, and built for scale from day one.
13
13
14
-
Our **serverless platform** handles the hard stuff: autoscaling reliable browser infrastructure, observability, and the messy details of web interaction so developers can focus on what their agents do, not how they do it. **Kernel makes AI tangible, practical, and powerful** — allowing developers to deploy agents that can truly interact with the digital world.
14
+
Our **serverless platform** handles the hard stuff: autoscaling reliable browser infrastructure, observability, and enabling human-in-the-loop interactions in production.
15
15
16
-
**If you're interested in building critical infrastructure for the next generation of AI applications, we'd love to chat.** One line of code deploys any web agent to our cloud. The rest is up to you.
16
+
**If you're interested in building critical infrastructure for the next generation of AI applications, we'd love to chat.**
17
17
18
18
## About the role
19
19
@@ -44,6 +44,7 @@ As **Founding Backend Engineer** at Kernel, you'll architect and build the core
44
44
-**$1k / mo budget** for LLM usage fees #vibecheck
- 💻 **Skills: Bun / Typescript, Go, Stainless, MySQL (PlanetScale)**
9
9
10
10
## About Kernel
11
11
12
-
We're building *the* developer platform that lets AI agents use real applications, starting with browsers. Our edge is an infrastructure platform that's extensible, observable, and built for scale from day one. We've shipped foundational tech, landed paying customers, and are well-funded while still in stealth.
12
+
We're building *the* developer platform that lets AI agents use applications, starting with browsers. Our edge is an infrastructure platform that's extensible, observable, and built for scale from day one.
13
13
14
-
Our **serverless platform** handles the hard stuff: autoscaling reliable browser infrastructure, observability, and the messy details of web interaction so developers can focus on what their agents do, not how they do it. **Kernel makes AI tangible, practical, and powerful** — allowing developers to deploy agents that can truly interact with the digital world.
14
+
Our **serverless platform** handles the hard stuff: autoscaling reliable browser infrastructure, observability, and enabling human-in-the-loop interactions in production.
15
15
16
-
**If you're interested in building critical infrastructure for the next generation of AI applications, we'd love to chat.** One line of code deploys any web agent to our cloud. The rest is up to you.
16
+
**If you're interested in building critical infrastructure for the next generation of AI applications, we'd love to chat.**
17
17
18
18
## About the role
19
19
@@ -42,6 +42,7 @@ As **New Grad Software Engineer** at Kernel, you'll build the core systems that
42
42
-**$1k / mo budget** for LLM usage fees #vibecheck
We're building *the* developer platform that lets AI agents use real applications, starting with browsers. Our edge is an infrastructure platform that's extensible, observable, and built for scale from day one. We've shipped foundational tech, landed paying customers, and are well-funded while still in stealth.
12
+
We're building *the* developer platform that lets AI agents use applications, starting with browsers. Our edge is an infrastructure platform that's extensible, observable, and built for scale from day one.
13
13
14
-
Our **serverless platform** handles the hard stuff: autoscaling reliable browser infrastructure, observability, and the messy details of web interaction so developers can focus on what their agents do, not how they do it. **Kernel makes AI tangible, practical, and powerful** — allowing developers to deploy agents that can truly interact with the digital world.
14
+
Our **serverless platform** handles the hard stuff: autoscaling reliable browser infrastructure, observability, and enabling human-in-the-loop interactions in production.
15
15
16
-
**If you're interested in building critical infrastructure for the next generation of AI applications, we'd love to chat.** One line of code deploys any web agent to our cloud. The rest is up to you.
16
+
**If you're interested in building critical infrastructure for the next generation of AI applications, we'd love to chat.**
17
17
18
18
## About the role
19
19
@@ -45,6 +45,7 @@ As **Founding Infrastructure Engineer** at Kernel, you'll architect and build th
45
45
-**$1k / mo budget** for LLM usage fees #vibecheck
Copy file name to clipboardExpand all lines: careers/intro.mdx
+3-3Lines changed: 3 additions & 3 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -4,11 +4,11 @@ title: "Join Us!"
4
4
5
5
## About Kernel
6
6
7
-
We're building *the* developer platform that lets AI agents use real applications, starting with browsers. Our edge is an infrastructure platform that's extensible, observable, and built for scale from day one. We've shipped foundational tech, landed paying customers, and are well-funded while still in stealth.
7
+
We're building *the* developer platform that lets AI agents use applications, starting with browsers. Our edge is an infrastructure platform that's extensible, observable, and built for scale from day one.
8
8
9
-
Our **serverless platform** handles the hard stuff: autoscaling reliable browser infrastructure, observability, and the messy details of web interaction so developers can focus on what their agents do, not how they do it. **Kernel makes AI tangible, practical, and powerful** — allowing developers to deploy agents that can truly interact with the digital world.
9
+
Our **serverless platform** handles the hard stuff: autoscaling reliable browser infrastructure, observability, and enabling human-in-the-loop interactions in production.
10
10
11
-
**If you're interested in building critical infrastructure for the next generation of AI applications, we'd love to chat.** One line of code deploys any web agent to our cloud. The rest is up to you.
11
+
**If you're interested in building critical infrastructure for the next generation of AI applications, we'd love to chat.**
Copy file name to clipboardExpand all lines: introduction.mdx
+32-14Lines changed: 32 additions & 14 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -5,22 +5,40 @@ description: "Infrastructure as-a-service for browser automations, web agents, a
5
5
6
6
Kernel is a powerful serverless platform that enables developers to deploy and scale browser automations without worrying about infrastructure.
7
7
8
-
## Why use Kernel?
8
+
-**Integrated browser environment** - Seamlessly create and connect your web agent to our cloud-based browsers on-demand
9
+
-**Parallel scaling** - Run concurrent browsers on-demand; we scale with you
10
+
-**Securely isolated** - Kernel browsers are sandboxed in individual isolated virtual machines
11
+
-**State persistence** - Maintain browser sessions across invocations, preserving cookies, auth tokens, and history
12
+
-**Live debugging** - View and interact with browser sessions in real-time using our Live View
13
+
-**Replays** - Review past browser session videos
14
+
-**Simple, predictable pricing** - Pay only for the resources you use without pre-defining infrastructure requirements
9
15
10
-
-**Local to production in seconds** - No configurations, no production pipelines. Deploy and run your code on our platform in the same amount of time it takes to type `bun run dev`.
11
-
-**Turn any agent into an API** - Kernel's platform turns every agent you write into an API that you can invoke from anywhere.
12
-
-**Parallel scaling** - Run thousands of concurrent browsers on-demand; we scale with you.
13
-
-**Securely isolated, easily observable** - Your code runs in isolated virtual machines, and we provide the tools you need to inspect and debug them.
14
-
-**Simple, predictable pricing** - Pay only for the resources you use, without pre-defining infrastructure requirements.
16
+
## How to use Kernel
15
17
16
-
## Browser support
18
+
There are two ways to use the Kernel platform:
19
+
### 1. Deploy and invoke your web agent on our platform
17
20
18
-
Our platform provides first-class support for browser-based workloads. For developers building web automations, AI agents, or any browser-dependent application, Kernel offers specialized capabilities:
21
+
Kernel's platform allows you to to [deploy](/launch/deploy) your app's code to our cloud. This allows you to co-locate your browser automation or web agent directly alongside the browser it controls in the cloud, minimizing latency and connection issues common with other platforms.
19
22
20
-
-**Integrated browser environment** - Seamlessly create and interact with our cloud-based browsers on-demand
21
-
-**State persistence** - Maintain browser sessions across invocations, preserving cookies, auth tokens, and history
22
-
-**Standby mode** - Restart browser instances in 20 milliseconds or less
23
-
-**Live debugging** - View and interact with browser sessions in real-time
24
-
-**Replays** - Review past browser sessions for debugging and analysis
23
+
Follow our [quickstart guide](/quickstart) to get started.
24
+
25
+
### 2. Connect to remote browsers via Chrome DevTools Protocol
26
+
27
+
For developers who already have browsers automations in production but want to use Kernel's cloud-based browsers, you can quickly spin up [remote browsers](/build/remote-browsers) and connect to them with the Chrome DevTools Protocol:
Our co-located architecture places your application code directly alongside the browser in the cloud, minimizing latency and connection issues common with other platforms. The best part is that all of this happens without you needing to configure Kubernetes, manage Docker containers, or wrangle AWS. We handle the infrastructure.
0 commit comments