Sure! Here's your project description rewritten in Markdown using the formatting tips you shared:
๐ง IntelCraft: Your AI Outreach Agent
๐ What is IntelCraft?
IntelCraft is your AI-powered research buddy, built to help students, freshers, and job seekers learn about companies quickly and write personalized emails โ whether you're applying for an internship, job, or preparing for an interview.
Instead of spending hours Googling and drafting cover letters, you get everything done in minutes โ thanks to real-time web scraping, smart AI summarization, and insightful reporting.
๐ฏ Why We Built This
As students ourselves, we found it frustrating and time-consuming to:
- Understand a companyโs background and goals
- Write compelling emails or cover letters
- Prepare for interviews with real insights
- Handle last-minute applications
Most existing tools are made for professionals โ not freshers or students.
So, we built IntelCraft to:
- Provide instant, accurate company insights
- Save hours of manual research
- Generate custom emails and outreach drafts
- Make job prep feel more strategic and less stressful
๐ง How It Works
Search for a Company Enter a company name โ we collect data from reliable online sources.
AI Breaks it Down Our AI summarizes:
- Company mission & values
- Latest news & achievements
- Competitors
- Opportunities & trends
- Get a Personalized Email Based on your interest + the companyโs insights, we auto-generate a thoughtful email or cover letter.
๐จโ๐ป Technologies We Used
- Frontend: React + TypeScript
- Backend: FastAPI + PostgreSQL
- Auth: Google OAuth
- AI Models: OpenAI, Gemini, Tavily
- Real-Time Updates: WebSockets
- Agent Flow: LangGraph for managing AI steps
๐ก What We Learned
- Debugging OAuth callback issues across local and deployed environments
- Optimizing AI responses with fallback logic
- Maintaining state and live updates through WebSockets
- Choosing the right AI provider for different use-cases
๐ ๏ธ Code Highlights
# Our AI Pipeline:
class ResearchPipeline:
def __init__(self):
self.collector = WebDataCollector()
self.analyzer = AIAnalyzer()
self.synthesizer = ReportGenerator()
async def process_company(self, company_name: str):
raw_data = await self.collector.gather(company_name)
insights = await self.analyzer.process(raw_data)
report = await self.synthesizer.create(insights)
return report
// WebSocket gives live updates while processing:
const ws = new WebSocket(`${WS_BASE_URL}/ws/${jobId}`);
ws.onmessage = (event) => updateResearchStatus(JSON.parse(event.data));
๐งโ๐ Who This Is For
- Students preparing for interviews
- Freshers applying to companies
- Anyone needing help with email drafts
- People curious about company background
โHi [Company], I admire your recent work on [X]. I'd love to contribute...โ โ This is generated instantly for you.
๐ Features
- โ Google Sign-In
- โ Company insight dashboard
- โ Personalized email/cover letter generation
- โ Real-time progress updates
- โ Clean & mobile-friendly interface
๐ What We Achieved
- โฑ๏ธ Cut research time by 90%
- ๐ฉ Auto-generated professional outreach messages
- ๐ Real-time research feedback loop
๐ Full-stack deployment with:
๐ฎ What's Next
- Resume parsing + matching company profile
- Comparison dashboard
- More AI models for diversity
- Export to PDF / Excel
๐ Final Thoughts
IntelCraft started at a hackathon โ now it's solving real problems.
If you're a student, fresher, or job seeker who struggles with company research or writing emails, we built this for you.
Built With
- fastapi
- framer
- gemini
- google-oauth
- langgarph
- neon
- openiai-api
- postgresql
- python
- react
- react-router
- tailwind
- tavily-api
- typescript
- unvicorn
- vite
Log in or sign up for Devpost to join the conversation.