AI Copilots
Real-Time Search for AI Copilots
AI copilots need to recall relevant context instantly. Moss runs semantic search directly in the browser via WebAssembly, so your copilot retrieves context without sending data to a third-party server.
> Query: webhook retry policy
Searching internal sources...6ms
Source
retry-policy.md
webhook-handler.ts
5 attempts, exponential backoff, capped at 24h
retryDelay: [1m, 5m, 30m, 2h, 12h]
100%
DATA STAYS ON-DEVICE
<10ms
IN-BROWSER RETRIEVAL
0
NO THIRD-PARTY DATA TRANSFERS
AI copilots need to recall relevant context instantly.
Moss runs semantic search directly in the browser via WebAssembly, so your copilot retrieves context without sending data to a third-party server.
The Problem
Copilots without context are just chatbots
An AI copilot that cannot recall your documentation, past conversations, or domain knowledge is just a generic chatbot with a different label. Traditional RAG pipelines require sending user queries to a cloud vector database, adding latency and creating a privacy liability. Every query leaves the user's device, hits a third-party server, and comes back. For enterprise copilots handling sensitive data, that architecture is a non-starter. You need retrieval that runs where the copilot runs.
An AI copilot that cannot recall your documentation, past conversations, or domain knowledge is just a generic chatbot with a different label. Traditional RAG pipelines require sending user queries to a cloud vector database, adding latency and creating a privacy liability. Every query leaves the user's device, hits a third-party server, and comes back. For enterprise copilots handling sensitive data, that architecture is a non-starter. You need retrieval that runs where the copilot runs.
How Moss Solves This
1
In-browser search via WebAssembly
Moss compiles to WebAssembly and runs directly in the browser. Your copilot retrieves context locally without any server round-trip.
2
Private by architecture
User queries never leave the device. No data is sent to third-party servers. Compliance-friendly for healthcare (HIPAA) and enterprise (SOC2) environments.
3
Hybrid search for precision
Combine semantic similarity with keyword matching. The alpha parameter lets you tune the blend for your specific domain and data.
Ship real-time retrieval in minutes
from moss import MossClient
client = MossClient(PROJECT_ID, PROJECT_KEY)
docs = [{ "text": "How do I track my order?" }]
await client.add_docs("my-index", docs)Frequently asked questions
Ready to ship faster AI products?
Moss gives you production-ready semantic retrieval without infrastructure complexity.