back

Auggy

An AI-powered workplace assistant for everyday work.

01

Overview

Auggy is an AI-powered workplace assistant that combines a desktop application, specialized AI agents, organizational knowledge retrieval, and workplace-tool integrations to simplify everyday employee tasks.

02

Achievements

  1. 01

    Multi-agent orchestration

    Designed and implemented LangGraph-based orchestration for multiple specialized workplace agents, coordinating task execution across integrated systems.

  2. 02

    Country-aware RAG

    Built a country-aware RAG system that filtered organizational knowledge by document metadata to produce contextually appropriate responses.

  3. 03

    Workplace agents

    Developed specialized agents for Jira, Google Calendar, Gmail, and PaddyField, extending the assistant into core workplace workflows.

  4. 04

    Productivity features

    Delivered employee productivity features including automated desktop-based time tracking and goal-setting with personalized 6–12 month learning roadmaps.

  5. 05

    Role-based access control

    Implemented Role Based Access Control (RBAC) to enforce precise user permissions, dynamically allowing or restricting actions and features based on user roles.

  6. 06

    Automatic time tracking

    Implemented automated time tracking, logging login/logout times as soon as the system boots up.

03

Architecture

Interactive: pick an option and tap a step to explore

Multi-agent orchestration

Sending asSigned in as
Multi-agent orchestrationIsolated user sessions feed a LangGraph orchestrator. A role check decides whether the user may reach each agent and returns an access-denied error otherwise. The knowledge agent runs RAG over Vertex AI Search with a country filter on document metadata.User A sessionOwn memory per userUser B sessionOwn memory per userOrchestratorBuilt with LangGraphRole checkIs this agent allowed?Access deniedError sent to userJiraTicketsCalendarMeetingsGmailEmailsPaddyFieldTime logsKnowledgeCompany docsHRRestrictedKnowledge agent: RAGFilter by countryDocument metadataVertex AI SearchInternal documentsGrounded answerBack to orchestrator

Every user gets a separate chat session with its own memory. A follow-up like "move it to Friday" resolves against that user's earlier messages, and conversations from different users never mix.

04

Timeline

CodingSpace IndiaAugment · 2022 – PresentAuggy
1 year
On this project
2024 – 2025
Active
Augment
Where

05

The stack

  • Electron
  • React
  • Redux
  • Tailwind CSS
  • FastAPI
  • RAG
  • Cron
  • LangGraph
  • LangChain
  • LangSmith
  • Vertex AI Search
  • OpenAI
  • Google Gemini
  • Redis
  • Docker
  • GCP
  • Vite
  • Vitest
  • Firestore
  • GitHub Workflows
  • GitHub Actions

Auggy

AI Assistant

⌘ K

Recent

Try a card, a recent request, or type your own.