RyAssist Platform
RyAssist is built from seven integrated systems that share context, data, and permissions. Each module works independently and more powerfully when used together.
Open WorkspaceConversations with full context and autonomous execution.
The AI Workspace is the primary interface for interacting with AI across your entire organization. Conversations maintain context across sessions, reference files and projects, and can trigger agent tasks directly from the chat interface. Streaming responses, artifact generation, and multi-modal input are built in.
Streaming responses with <200ms time-to-first-token
Cross-session context and memory
Artifact generation with real-time preview
File attachment and document parsing
Shareable conversation threads
Autonomous agents that collaborate and execute in parallel.
Each agent is a persistent entity with its own memory, tools, and objectives. Agents communicate through a shared message bus, can delegate subtasks to other agents, and operate concurrently without blocking your workflow. A supervisor agent coordinates complex tasks that span multiple specialized agents.
Persistent agent memory across sessions
Tool-calling with type-safe schemas
Agent-to-agent delegation
Parallel execution with configurable concurrency
Audit trail for every agent action
Visual workflow builder for multi-step automation.
The Agent Canvas lets you construct autonomous pipelines visually. Drag-and-drop nodes representing retrieval, reasoning, transformation, and action steps. Connect them with typed edges, configure conditions and branches, and deploy as a reusable workflow that runs on a schedule or responds to triggers.
Node-based visual workflow editor
Conditional branching and loops
HTTP, webhook, and event triggers
Reusable sub-workflow components
One-click deploy to production
AI-native storage with semantic search across everything.
Files in RyAssist are first-class objects. Upload documents, images, code, and data — they are indexed for AI retrieval, versioned automatically, and shareable at the user, project, or organization level. The AI Workspace can reference any stored file without explicit retrieval prompts.
Automatic semantic indexing on upload
Full-text and vector hybrid search
File versioning with diff viewer
Per-file access control and expiring links
Preview for PDF, code, images, and structured data
Kanban, assignments, and AI assistance across every task.
Projects in RyAssist are work containers — they scope files, conversations, and agents to a defined objective. Within a project, tasks can be assigned to humans or AI agents, tracked on a kanban board, given due dates, and linked to the conversations and documents they produced.
Kanban and list views with custom fields
AI agent task assignments
Due dates, priorities, and dependencies
Project-scoped file and conversation context
Activity feed per project
Multi-tenant architecture with granular access control.
Organizations are the top-level container for teams and enterprise customers. Each organization has its own user directory, roles, and billing. Workspaces within an organization can be scoped to teams, departments, or projects, with inherited or overridden permissions at each level.
Nested workspace hierarchy
Custom roles with field-level permissions
Cross-organization collaboration with guest access
Organization-wide AI usage analytics
Per-organization model and data policies
Zero-trust authentication with enterprise SSO.
Authentication in RyAssist uses OAuth 2.0 and supports SAML 2.0 and OIDC for enterprise SSO. Every request is authenticated and authorized at the resource level. All data is encrypted at rest and in transit, with full audit logging of every access event and agent action across the platform.
SAML 2.0 and OIDC enterprise SSO
Per-resource OAuth 2.0 scopes
AES-256 encryption at rest, TLS 1.3 in transit
Immutable audit log with SIEM export
SOC 2 Type II architecture
Every module in RyAssist shares a common data layer. A file uploaded in Cloud Files is immediately available as context in AI Workspace conversations, retrievable by agents on the Canvas, and linkable from Projects. No integrations, no sync, no silos.