# Using Kernel AI

Get AI assistance for genetic design tasks and navigation.

## What is Kernel AI?

Kernel AI is an AI assistant that helps you:

* Navigate the Kernel interface
* Answer genetic design questions
* Find constructs and parts
* Explain biological concepts
* Assist with design decisions

## Accessing Kernel AI

### Sidebar

Click **Kernel AI** in the left sidebar to open the chat interface.

![Kernel AI Interface](/files/CfB3LmEn8qlszEwocZy8)

### In-Editor

Click **AI** in the construct editor's right sidebar for context-aware assistance.

### Keyboard Shortcut

Press **Cmd/Ctrl + K** to quickly access Kernel AI.

## What You Can Ask

### Navigation

Ask Kernel AI to help you find things:

* "Take me to my recent constructs"
* "Find all promoters in my workspace"
* "Show me the CMV constructs"
* "Open the shared parts repository"

### Design Questions

Get help with genetic design:

* "What promoter should I use for high expression in CHO cells?"
* "Explain the difference between SV40 and BGH terminators"
* "How do I optimize this sequence for E. coli?"
* "What signal peptide works best for secretion?"

### Construct Information

Ask about your designs:

* "What parts are in this construct?"
* "Show me the sequence of the heavy chain"
* "What's the GC content of this region?"
* "Are there any HindIII sites?"

### Workflow Help

Get guidance on tasks:

* "How do I create a new construct?"
* "How do I share a repository?"
* "How do I run a simulation?"
* "How do I export as GenBank?"

## Example Conversations

### Finding Parts

**You**: Find all T7 promoters in my workspace

**Kernel AI**: I found 3 T7 promoter parts in your workspace:

1. T7-Promoter (in Team Parts)
2. T7-Strong-Promoter (in Expression Library)
3. T7-Lac-Promoter (in Inducible Parts)

Would you like me to open any of these?

### Design Advice

**You**: What's the best way to express an antibody in HEK293?

**Kernel AI**: For antibody expression in HEK293 cells, I recommend:

* CMV or EF1α promoter for strong expression
* IgK or native signal peptide for secretion
* BGH poly(A) for 3' UTR
* Consider dual-plasmid system for HC + LC

Would you like me to help you set up a construct with these elements?

### Navigation

**You**: Take me to the construct I was working on yesterday

**Kernel AI**: Based on your recent activity, you were working on:

* HC-Expression-v3 (modified 18 hours ago)
* LC-Optimized (modified 20 hours ago)

Which one would you like to open?

## Tips for Effective Use

### Be Specific

More specific questions get better answers:

* Instead of "help with my construct"
* Try "help me add a His-tag to the C-terminus of my protein"

### Provide Context

Include relevant details:

* The construct you're working on
* Your expression system
* What you're trying to achieve

### Ask Follow-ups

Continue the conversation:

* Ask for more details
* Request alternatives
* Get step-by-step guidance

## Capabilities

### What Kernel AI Can Do

* Search and navigate workspace content
* Explain genetic concepts
* Recommend parts and designs
* Guide you through workflows
* Answer questions about Kernel features

### Limitations

* Cannot modify constructs directly (you make the changes)
* Answers based on general knowledge and your workspace
* Complex designs may need expert review
* Experimental predictions should be validated

## Best Practices

### Design Assistance

* Use AI for initial guidance
* Verify recommendations
* Consider multiple approaches
* Validate critical designs

### Learning

* Ask "why" questions to learn
* Request explanations of concepts
* Explore alternatives

## Next Steps

* [Quick Start Guide](/getting-started/quick-start.md) — Get started with Kernel
* [Creating Constructs](/constructs/creating-constructs.md) — Design with AI help
* [Part Library](/parts/part-library.md) — Find parts AI recommends


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.kernel.asimov.com/kernel-ai/using-kernel-ai.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.