Inspiration
Modern work is scattered across too many tools. Important context lives in GitHub, Notion, Slack, calendars, documents, and old conversations. To answer a simple question like “What did I miss today?”, people often have to open five different apps, search manually, and piece the story together themselves.
We wanted to build something that feels simpler and more natural: not another dashboard, but a teammate you can text.
Context was inspired by the feeling of messaging a thoughtful coworker who already knows the project, remembers what happened, and can quickly tell you what matters.
What it does
Context is a native iOS chat experience for understanding your workspace.
Users can ask questions like:
What did I miss today?
What changed in the repo?
Which PRs need attention?
What should I work on first?
What decisions were made in Notion?
Context gathers information from connected tools, reasons across them, and replies in a concise conversational style. Responses include source cards so users can see where the answer came from, such as GitHub, Notion, or Calendar.
The app also supports a daily brief, where Context proactively sends a morning summary of important updates without requiring the user to ask first.
The core idea is:
You don’t search your workspace anymore. You text it.
How we built it
We designed Context as a polished native iOS app using SwiftUI, async/await, MVVM architecture, native animations, dark mode, streaming responses, and a minimal chat-first interface inspired by Apple Messages.
The backend is built with Python and FastAPI. It handles AI orchestration, authentication, prompt management, memory, and integrations with external services like GitHub and Notion.
We used Supabase for user accounts, conversation history, workspace connections, and cached summaries.
For AI, we used the OpenAI Responses API with tool calling, structured outputs, retrieval, memory, and streaming. Codex helped us rapidly plan, structure, and iterate on the product, while GPT-5.6 powered the teammate-like reasoning layer that turns scattered workspace data into useful conversational answers.
Challenges we ran into
The hardest challenge was keeping the product simple.
It is easy for a workspace AI product to become a dashboard, command center, or analytics tool. We intentionally avoided that. Every feature had to pass one test:
Does this make Context feel more like a teammate?
Another challenge was designing responses that feel human and useful without becoming verbose. Context should not sound like a generic chatbot. It should sound like a professional coworker who knows what matters and gets to the point.
We also had to think carefully about source transparency. Since Context reasons across multiple tools, we wanted every important answer to show where the information came from without cluttering the conversation.
Accomplishments that we're proud of
We are proud of creating a product concept that feels focused, memorable, and emotionally clear.
Instead of building a broad productivity platform, we built around one strong interaction: texting your company.
We are also proud of the conversational design. Context does not just summarize information. It prioritizes, explains, and responds in a tone that feels helpful, calm, and coworker-like.
The demo story is simple enough to understand immediately:
A user opens the app, sees a morning brief, asks what to work on first, and gets a prioritized answer backed by real sources.
What we learned
We learned that the interface matters as much as the model.
A powerful AI system can still feel overwhelming if it is wrapped in a complicated product. By making the app feel like a normal conversation, the AI becomes easier to trust and easier to use.
We also learned that source-backed answers are essential for workspace AI. Users do not just want an answer. They want to know whether the answer is grounded in GitHub, Notion, Calendar, or another real source.
Most importantly, we learned that the best AI teammate is not the one that says the most. It is the one that knows what matters.
What's next for Context
Next, we want to expand Context into more workplace tools while keeping the same simple chat-first experience.
Planned next steps include:
Slack integration
Gmail integration
Calendar integration
Push notifications
Voice conversations
Action execution, such as commenting on PRs or creating follow-up tasks
Multiple workspace support
Deeper long-term memory across projects and conversations
The long-term vision is for Context to become the first place people go when they need to understand what is happening at work.
Log in or sign up for Devpost to join the conversation.