Your idea. Your AI.
A working preview.
Build a web idea, test a component or explore a new interface in one focused desktop workspace. Keep AI chat, your code and live preview together. Choose your AI, point to what you want changed, and keep building.

You choose the model your key provides. Switch models anytime from the sidebar without losing conversation state.
Clean code editor with live rendering. Click any element in preview to leave point-to-edit notes for the AI.
Context management, supported prompt caching and optional smart routing help manage AI usage. Savings vary by task, model and provider.
Multi-Provider Support
One workspace for all your AI providers.
Connect your accounts and switch anytime. Turn on optional Smart Model Routing in Settings to automatically route simpler tasks to cheaper companion models.
A place for your next idea
Try the idea.
Stay in the flow.
For developers testing a component, designers exploring an interface, curious makers prototyping a page and people learning by building. Keep the conversation beside the result, from the first prompt to the next refinement.
The idea can lose momentum
- Copy, paste, switch, repeat: Moving generated code between a chat, editor and browser interrupts a quick experiment.
- Setup becomes the project: Exploring optional extensions and integrations can take time away from the thing you wanted to build.
- Feedback loses its reference: Describing which button, card or heading needs changing adds another explanation between you and the result.
- Your attention gets divided: More windows and settings compete with the idea you came to test.
One window. Keep building.
- See the result beside the chat: Keep the conversation, code editor and running preview together while you build and refine.
- Focus on the next change: Select elements in the preview, add notes and send those visual references to the AI.
- Your AI, your choice: Bring your own API key for a supported provider, or connect compatible local models through Ollama.
- Keep AI usage visible: Token tracking, context optimization and optional smart routing help manage usage. Savings vary by task, model and provider.
- Add voice when it suits you: Optional local voice components download through Pulse, which handles their configuration in the background.
Less to assemble. More room to focus.
No MCP servers, plugin marketplace or skills to assemble for the core chat–code–preview workflow. Choose a supported provider and your key, or set up Ollama with a compatible local model. Your workspace stays on your computer; cloud AI usage is separate.
Point, pin, and edit in the live preview.
No more copying back and forth or explaining which button is broken. Pulse includes a built-in code editor and live browser preview side-by-side with chat.
Visual Inspector Crosshair
Click the crosshair directly over your running preview to inspect any UI element.
Multi-Element Pinning
Pin as many buttons, headings, or cards as you need. Type notes on each simultaneously.
One-Shot AI Execution
Click Send to AI — Pulse bundles your visual references and updates the code in place.
Natural voice mode.
Zero setup. Just talk.
Add voice without assembling a separate speech stack. Pulse downloads and configures its local voice components in the background, so you can stay in the workspace.
Pulse quietly downloads the lightweight Kokoro TTS model (~92 MB) and dictation engine in the background, wires up the local backend pipeline automatically, and sets everything up. You do zero configuration — simply pick a voice from the menu and start talking.
- Zero Python, zero terminal: Native background configuration with zero user intervention.
- Bidirectional local audio: Whisper for voice input + Kokoro for natural voice output.
- Local voice processing: Speech input and playback use local voice components. Sending the resulting text to cloud AI still uses your chosen provider.
Built-in token optimization
Manage long conversations.
Use context with care.
Long conversations can include more context than a task needs. Pulse uses context management and optional smart routing to help manage usage. Actual costs depend on your task, model and provider.
Context Selection
Only the relevant turns and snippets needed for the current question are sent to the model, ignoring idle context.
History Compaction
Older messages are condensed into high-density semantic summaries instead of ballooning the request token payload.
Code Skeletonization
Large source files are summarized to structural signatures (classes, functions, types) when full details are redundant.
Prompt-Cache Reuse
Uses supported provider prompt caching when eligible. Availability, cache hits and discounted rates depend on the provider and request.
Response Caching
Eligible repeated requests can reuse local cached responses. Requests that need a new model response still incur provider usage.
Smart Model Routing
Turn on in Settings to automatically route simpler tasks and minor edits to faster, lower-cost companion models in the same provider family.
Your workspace stays local.
Cloud requests go to your provider.
Chats and workspace files are stored on your computer. When you use cloud AI, the prompt and included context are sent to the selected provider under its own data policies. Compatible Ollama models run locally after setup.
Honest Software
Own the workspace. Choose your AI.
$19 once for the Pulse software. Cloud AI usage is billed separately by your provider; local models require suitable hardware and setup.
macOS & Windows included
Instant download after purchase · Direct email support
Questions & Answers
Frequently asked questions.
Craft software for people who care about details.
Pulse is the flagship desktop project from builtbyangelo.co. More tools, shaders, and experimental local AI workspaces are in active development.