AI chatbot that helps you keep a brag document about your achievements at work.
Helps employees shine in their performance reviews by keeping a record of their accomplishments, and crafting beautiful documents to share with their managers for the year-end review, semi-annual review, or any other time period. Or even just a "State of me" weekly update to manager, and monthly update to skip manager.
The existing UI is already a familiar chatbot experience, and has a "Canvas" mode that allows the human and an LLM to collaborate on a single document. There is an existing data model that represents these Documents, as well as other things.
Day to day, the user is just telling the chat bot about things they've done at work. Tasks they've completed, potentially things that are blocking them, unexpected events, etc. The chatbot will use an LLM to populate a database via the Achievement model - a Achievement is being a thing that the user has achieved. Users might "brag" to the bot about one thing they've just achieved, or tell the bot about a bunch of things they've done over the last few weeks and didn't have time to brag about yet.
From time to time, the user will want a summary of these achievements over some period. The period could be 6-12 months for a performance review, one week for a weekly summary to manager, one month for a monthly summary to skip manager, etc. The chatbot will use an LLM to generate a summary of these brags for the user, presenting it as a Document in the "Canvas" mode.
- Interactive chatbot interface for logging achievements
- Canvas mode for collaborative document editing with AI
- Intelligent achievement tracking and categorization
- Multi-company support as you change employers
- Email integration for sending and receiving achievements
Documents can be created and edited inside the UI with Canvas mode. The user can then just copy the content and paste it somewhere else, like in an email to their manager.
There are also various publishing options:
- After creating an update Document, user can:
- publish it to a url like https://bragdoc.ai/edspencer/world-labs/weekly/some-private-fragment/2024-12-20. Allows manager to view the doc. The some-private-fragment is a password that lets the manager view the document. Should be generated randomly from a dictionary of English words
- Update a Google Document by creating a new page and adding the document content to it
- Email the document to a recipient (would allow user to schedule a weekly update report generation, with it being emailed to them, their manager or both)
The document editing UI is already complete.
- Record individual or batch accomplishments
- Track projects and collaborators
- Document blockers and unexpected challenges
- Capture daily/weekly tasks and their impact
- Performance review documents (6-12 month summaries)
- Weekly updates for direct managers
- Monthly summaries for skip-level managers
- Custom time period reports
- AI-assisted document editing and refinement
- Email integration for sending achievements directly
- Automated reminder system for regular updates
- Email responses with AI-processed achievement logging
- Proactive notifications for extended periods without updates
- Persistent storage of achievements and documents
- Secure multi-company data separation
- Historical achievement tracking and access
- Email/password authentication
- Google authentication
- GitHub authentication
- Automatically creates Achievements out of connected repositories!
- Can retro-actively link a repo, creating a Project for it and filling it with Achievements from the repo history [Premium Feature]
The application now supports bi-directional email communication:
- Send achievements directly to
hello@bragdoc.ai - AI automatically extracts achievements from email content
- Supports rich text and plain text email formats
- Achievements are linked to your user account and companies/projects
- Weekly/monthly achievement summaries
- Performance review documents
- Manager updates
- Customizable email templates using React Email
- Unsubscribe management
- Email sender verification
- Mailgun webhook signature validation
- Rate limiting and spam protection
- GDPR-compliant data handling
Should have some copy about how many things you will forget without it. Sign up / Sign in buttons
- Use LLM to generate a bunch of past conversations that a user could have had with Braggart to save their work on something
- Use this to drive Evals
- Use it to generate test & dev data (needs to be spread out over last 12 months ideally)
The project is organized into several key areas:
app/(marketing): Marketing pages and layouts- Landing page with feature highlights
- SEO-optimized metadata
- Dark mode support
- Responsive design
app/(auth): Authentication flowsapp/(chat): Main application interface- Chat interface for logging achievements
- Canvas mode for document collaboration
- Document generation and editing
components/marketing: Marketing page components- Hero section
- Feature showcases
- Pricing plans
- FAQ section
components/ui: Reusable UI components (based on shadcn/ui)
type Achievement = {
id: uuid,
userId: uuid,
createdAt: Date,
updatedAt: Date,
eventStart: Date, //detected start data of the Achievement
eventEnd: Date, //detected end date of the Achievement
eventDuration: string, //day, week, month, quarter, half year, year - how long a time period this Achievement covers (but eventStart/eventEnd are canonical)
userMessageId: uuid, //the text the user submitted about this Achievement saved here
summary: string, //the LLM summary of originalText
title: string, //bullet list-compatible title for the Achievement (LLM generated)
details: string, //most
};
type UserMessage = {
id: uuid,
originalText: string, //what the user actually said
achievements: Achievement[], //the one or more Achievements detected in what the user said
};- Free
- Some limitation on usage
- No integration with GitHub
- $3/month
- $30/year
- Integrates with one GitHub repository
- Unlimited Brags/Docs/everything
- $9/month
- $90/year
- Integrates with unlimited GitHub repositories
- Scheduled updates
- Publishing options
Should be integrated with Link or Stripe (Link has a really nice UX).
This application was initially built on top of a sample Next.js app template from Vercel. It has that following application-level features:
- Next.js App Router
- Advanced routing for seamless navigation and performance
- React Server Components (RSCs) and Server Actions for server-side rendering and increased performance
- AI SDK
- Unified API for generating text, structured objects, and tool calls with LLMs
- Hooks for building dynamic chat and generative user interfaces
- Supports OpenAI (default), Anthropic, Cohere, and other model providers
- shadcn/ui
- Styling with Tailwind CSS
- Component primitives from Radix UI for accessibility and flexibility
- Data Persistence
- Vercel Postgres powered by Neon for saving chat history and user data
- Vercel Blob for efficient file storage
- NextAuth.js
- Simple and secure authentication
This template ships with OpenAI gpt-4o as the default. However, with the AI SDK, you can switch LLM providers to OpenAI, Anthropic, Cohere, and many more with just a few lines of code.
You will need to use the environment variables defined in .env.example to run the Brag AI ChatBot. It's recommended you use Vercel Environment Variables for this, but a .env file is all that is necessary.
Note: You should not commit your
.envfile or it will expose secrets that will allow others to control access to your various OpenAI and authentication provider accounts.
- Install Vercel CLI:
npm i -g vercel - Link local instance with Vercel and GitHub accounts (creates
.verceldirectory):vercel link - Download your environment variables:
vercel env pull
pnpm install
pnpm devYour app template should now be running on localhost:3000.