Inspiration
As a product manager in the telecom industry, I witnessed how successful product launches depend on seamless collaboration across internal teams and external partners—from developers and handset providers to network suppliers and business stakeholders. Our matrix organization brought together diverse perspectives spanning hardware, software, networks, customer support, and business development. While these domains appeared disconnected on the surface, they were intricately linked through subtle dependencies critical to product success. Every planning session revealed the same frustration: teams spent valuable time revisiting previous discussions and re-explaining insights from scattered documents instead of advancing new ideas. What we desperately needed was an intelligent platform that could consolidate all our content—documents, meeting notes, technical specs—and automatically surface the critical connections and priorities hidden within. Imagine querying this collective knowledge through an AI agent that could instantly answer questions like "What are the key dependencies between our network architecture and handset compatibility?" or "Which issues require cross-organizational attention?" The technology to make this vision real simply didn't exist then. Today it does. OrgMind transforms collaborative knowledge work. Organizations can now create shared topic spaces where cross-functional teams and external partners upload documents, create content, and let AI automatically build knowledge graphs that reveal non-obvious connections and priorities. When the graph surfaces a critical insight, participants can immediately dive deeper through AI chat that draws from the entire document corpus. The applications extend far beyond cross-organizational projects. Internal teams can analyze research papers, market reports, and strategic documents to identify emerging trends invisible in individual sources. Legal teams or law firms can use it to upload contract agreements, case studies and other client documents and identify emerging trends and patterns in legal context to help clients address emerging risks or opportunities. Research teams can upload diverse articles and discover patterns that would remain hidden through traditional reading. Individual professionals can organize their knowledge domains and extract insights that accelerate decision-making. OrgMind makes organizational intelligence tangible, collaborative, and actionable.
What it does
OrgMind is a collaborative intelligence platform that transforms how teams share knowledge, discover insights, and make decisions together. Create Topic Graphs Users create topic graphs—shared repositories where teams can consolidate knowledge around projects, initiatives, or research areas. Content enters the graph through two channels: uploading documents (PDF, DOCX, HTML, and other popular formats) or creating content directly using our integrated Lexical rich-text editor. We recommend limiting each graph to approximately 30 documents for optimal performance. Invite Collaborators Each graph functions as a shared workspace with controlled access. Users invite team members from within their organization or external partners from the platform, with permissions managed through customizable roles. All participants can upload documents and create content, building a collective knowledge base in real-time. Discover Connections Through Knowledge Graphs When users click "Knowledge Graph," OrgMind automatically visualizes the hidden relationships within their content—displaying an interactive network of nodes and edges that reveal how concepts connect across documents. Clicking any node or edge surfaces detailed information about that relationship. Users can search for specific concepts and explore the surrounding subgraph, uncovering connections that would remain invisible through traditional document review. Chat with Your Knowledge OrgMind's AI chat interface enables deep exploration of any insight discovered in the knowledge graph. Unlike typical RAG solutions that offer a single conversation thread, OrgMind lets users create unlimited chat threads within each graph—each focused on a different topic, node cluster, or question. One thread might explore dependencies identified in the knowledge graph, another might investigate specific technical challenges, and yet another might synthesize findings across multiple documents. Each conversation is grounded exclusively in the graph's content through Retrieval Augmented Generation, ensuring answers are always traceable to source material. This graph-centric approach—where conversations, knowledge graphs, and documents are organized by topic—makes OrgMind uniquely powerful for focused collaboration around specific initiatives. Who Benefits OrgMind serves diverse collaboration scenarios: - Enterprise teams: Cross-functional project coordination, strategic planning, operational process documentation - External partnerships: Joint ventures, strategic alliances, B2B customer engagement, supplier collaboration - R&D and innovation: Multi-organizational research initiatives, competitive intelligence, market trend analysis - Individual researchers: Personal knowledge synthesis across geopolitics, economics, social issues—upload news articles, analyst reports, podcast transcripts, and research papers to discover emerging patterns - Research communities: Teams conducting collaborative investigations where multiple perspectives strengthen collective understanding Whether coordinating a complex product launch across partners, synthesizing market intelligence, or conducting deep research on emerging trends, OrgMind turns scattered documents into actionable organizational intelligence.
How we built it
Technology Stack
We built OrgMind on a modern, scalable architecture designed for enterprise reliability: - Frontend: TypeScript, Next.js, and React, deployed on Google Cloud Run's serverless platform for automatic scaling - Backend: Golang API server with PostgreSQL database, providing robust data management and type safety - Knowledge Graph: Zep cloud platform for creating and querying topic-based knowledge graphs, with each graph maintaining its own semantic network - Document Storage: AWS S3 for secure, scalable document and content storage - RAG & AI: Google's Gemini AI integrated through the GenAI SDK, leveraging File Search API for retrieval and Server-Sent Events (SSE) for streaming chat responses - Voice Intelligence: Gemini Live API with integrated Google Search capability for our AI agent The system architecture supports enterprise authentication through OpenID Connect, OAuth, and SAML integration, currently implementing credential-based authentication with SSO capabilities ready for deployment.
Modular Development Approach
We structured development around core modules: authentication, graph management, participant administration, document handling, knowledge graph visualization, and chat services. For each module, our team took an integrated approach—designing UI components, defining API contracts with JSON schemas, creating backend DTOs, and mapping domain objects to database tables simultaneously rather than sequentially. This parallel development strategy significantly reduced iteration cycles. By designing frontend and backend together with clear contracts, we minimized database migrations and avoided the typical integration friction of waterfall development. The result was faster delivery and more resilient data architecture.
AI-Powered Development with Kiro
Kiro became our AI development partner throughout the project. We began by creating a comprehensive steering document that established global rules covering technology stack, design principles, error handling patterns, coding styles for both TypeScript and Go, and architectural patterns for layered backend structure (routing/handler, service, repository layers). This 388-line steering document served as Kiro's constitution document throughout development. Our workflow with Kiro evolved through experimentation: we provided both high-level requirements and detailed system specifications—frontend, backend, and database design—in the same conversation. This holistic context proved critical for Kiro to understand system goals accurately. For each feature, Kiro generated specification documents with user acceptance criteria that captured requirements correctly over 80% of the time, requiring only minor iteration. What started as a modest project targeting basic text editing and knowledge graph generation expanded dramatically as we discovered Kiro's capabilities. Through iterative conversations and spec refinement, we ultimately consumed 400+ credits while building sophisticated features: multi-format document upload, full-featured chat interface, and end-to-end RAG implementation with Google's latest File Search API. The consistent architecture defined in our steering document made reviewing Kiro's generated code straightforward. While we occasionally identified bugs—typically related to newer libraries outside Kiro's training—the overall code structure adhered remarkably well to our architectural vision.
Challenges we ran into
Serverless Debugging
While serverless environments like Cloud Run streamline deployment, initial setup required significant effort, particularly debugging cloud build and runtime issues. Simple bugs—like configuration files not mounting in the correct directory—became tedious to debug without direct visibility correlating issues with logs. Failures related to ServiceAccount access and permission roles during build and runtime proved especially challenging due to indirect correlation between cloud build logs and the serverless environment.
API Integration Complexity
Using a wide range of API-based services introduced complexity in managing API keys as secrets and provide them as environment variables to the cloud run deployments. Ensuring secrets are set up prior to running cloud build added an additional layer dependency that we had to manually handle. Missing a secret or not mounting secrets, config file in the right directory led to unnecessary triggering of building. We plan to make the process more automated as we continue improving our deployment workflow.
Framework vs. Library Trade-offs
Our AI agent needed to run within the LiveKit Agent SDK framework. While most of the Gemini Live integration code was generated using Google AI Studio Build, a key challenge was adapting this generated code to fit seamlessly within the LiveKit SDK environment. Working within a framework inherently provides less flexibility compared to using AI SDKs directly (such as Google GenAI or OpenAI SDK). Throughout development, we continually balanced the trade-offs between leveraging the framework’s capabilities and writing custom code using libraries for greater control.
Accomplishments that we're proud of
Production-Ready Architecture
Rather than building a prototype to submit for hackathon, we rather took an approach to architect Meetspace for real-world deployment from day one. Our scalable serverless infrastructure on Google Cloud Run, combined with proper security implementations, managed secrets, and enterprise authentication support, means this project can transition directly to production use. We've built on solid foundations—proper database design, comprehensive API architecture, and cloud-native best practices—that will support growth from initial users to enterprise scale.
Solving Real Business Problems
We're most proud that OrgMind addresses genuine pain points we've experienced firsthand in business collaboration. By combining real-time AI assistance, intelligent meeting transcription, secure document sharing, and unified workspace management, we're tackling the fragmented, context-poor meeting experiences that plague modern organizations. We believe OrgMind can measurably improve productivity in business collaboration and communication, helping teams make better decisions with better context, whether they're internal cross-functional groups, external partnerships, or healthcare providers serving patients remotely.
AI-Powered Development Workflow
We established a development workflow that demonstrates the transformative potential of AI-assisted software engineering. Prior to this project, our team used a multi-tool approach: Claude Sonnet 4.5 for system design and requirements gathering, followed by Gemini CLI for agentic development. While effective, this pipeline required manual handoffs between tools and contexts. Kiro changed everything. By adopting Kiro's specification-driven approach, we automated our entire development workflow end-to-end. Kiro's deep repository understanding enabled it to maintain context across the full stack—from requirements through implementation—while its analytical capabilities helped us architect robust, modular specifications that remained easy to review and validate. The steering document became our force multiplier: once established, it ensured architectural consistency across hundreds of generated files while maintaining our coding standards and design patterns. This allowed us to focus human attention where it matters most—validating requirements, reviewing critical logic, and making strategic architectural decisions—while Kiro handled the repetitive, error-prone work of translating specifications into production code. The results speak for themselves: we accelerated development velocity, improved code quality through consistent patterns, and delivered more sophisticated features than initially scoped—all while maintaining architectural integrity. This workflow proves that thoughtful AI integration doesn't just speed up development; it elevates the entire software engineering process, allowing teams to operate at a higher level of abstraction while maintaining control over quality and design.
The intersection of these achievements—innovative development practices, production-ready engineering, and real business value—represents what we believe makes OrgMind not just a successful hackathon project, but the foundation of a product that can genuinely improve how people work together.
What we learned
Cloud Deployment
Deploying to a serverless environment presented challenges during initial deployment—errors surfaced at multiple stages: Dockerfiles, code itself, build time, container image creation, and ServiceAccount permissions. However, once deployment succeeded and our service went live with a public IP and domain, subsequent updates became remarkably efficient. Push code, trigger build, and everything's in production! Initially, we had concerns about debugging and log collection in serverless platforms like Cloud Run. Those doubts vanished quickly—GCP's logging support is advanced and feature-rich, proving that serverless platforms can scale effectively. Managed secrets and SSL certificates also eliminated the burden of manual certificate creation and renewal, as Google Cloud Platform handles these automatically. Collaboration & Documentation Modern API-based development feels like working with extended teams of developers from the organizations that provide APIs. When integrating APIs from LiveKit, OpenAI, or Gemini, we use the documentation, but the process feel like we are working in collaboration with those development teams. In today's cloud software environment, API developers are virtual team members—their work becomes part of our platform, making us feel like we're building one piece of a larger puzzle. Clear documentation is crucial. It directly impacts developer productivity, especially for complex integrations where our platform connects to dozens of third-party systems.
Agentic Development and Kiro
By adopting Kiro's specification-driven approach, we automated our entire development workflow end-to-end. Kiro's deep repository understanding enabled it to maintain context across the full stack—from requirements through implementation—while its analytical capabilities helped us architect robust, modular specifications that remained easy to review and validate.
Business Strategy
Business Model & Revenue Generation
Every great product solves a real customer problem. OrgMind is no exception. We invested time white boarding our target industries, user segments, their challenges, and how our product addresses those pain points. Identifying target customers is half the story; generating revenue is the other half. A product customers love won't survive without a solid profit model. For enterprise customers, we demonstrate ROI to justify pricing that reflects value. For individual users, pricing need to consider API token costs, cloud computing, third-party services, operational expenses and a the margin after covering all the costs to generate earnings for sustainable business growth.
What's next for OrgMind
For product/platform -
- Enhance the chat interface with agentic capabilities by integrating third-party tools, custom enterprise tools, and MCP servers. Provide an admin console that allows enterprise customers to configure and manage the tools used by their OrgMind AI agents
- Integrate voice-chat functionality, enabling users to interact through natural voice conversations connected to the topic context and underlying knowledge graph.
- Support Enterprise authentication and SSOs including OpenID, SAML, Okta, Google and other OpenID providers.
- Enhance team collaboration and real time messaging functionality, with more scalable caching infrastructure, Pub-Sub and message queue integration.
- Improve the overall system architecture to deliver OrgMind service at scale. ### For Go-To-Market
- lead generation, prospecting and running pilot with enterprise customers from target verticals.
- Content marketing - develop content for marketing and customer supports, content including blogs, user guides, videos, AI sales agent.
- Social media marketing.
- Explore business development and partnership opportunities with businesses whose products are complement with our offering.
Built With
- cloudrun
- gcp
- gemini
- golang
- nextjs
- postgresql
- s3
- tailwind
- zep
Log in or sign up for Devpost to join the conversation.