Project Info

C3

This project did not submit a demo video on Devpost.

Inspiration

Climate tech is a complex ecosystem of sectors, but career discovery within it is fragmented and unclear. Students and professionals want to work in climate but struggle to understand where their skills fit. We were inspired to bridge the gap between climate innovation and climate talent.

What it does

C3 is an AI-powered Climate Career Explorer that connects structured climate taxonomy with real job listings. It parses resumes, extracts skills, and semantically matches users to categorized climate roles. Users receive personalized job recommendations with percentage-based compatibility scores.

How we built it

We structured the Climate Tech Map dataset into a domain-specific skills and sector database. We scraped LinkedIn climate jobs using Bright Data and ingested them into Supabase. Using RAG, embeddings, GPT-4, and a Fetch.ai agent, we built a conversational job-matching system.

Challenges we ran into

Normalizing messy climate skill data into a clean taxonomy was difficult. Resumes vary significantly, making structured extraction challenging. Ensuring accurate retrieval without hallucination required careful RAG grounding.

Accomplishments we're proud of

We built an end-to-end climate-focused AI job intelligence pipeline. We successfully connected structured innovation data with real job listings. We implemented transparent resume-to-job percentage matching.

What we learned

Structured domain data dramatically improves AI retrieval quality. Semantic matching is far more effective than keyword-based job searches. AI systems are most powerful when grounded in clean, explainable data models.

What's next

for Untitled We plan to enable real-time job updates and skill gap recommendations. We aim to expand into employer dashboards and personalized learning pathways. Our long-term vision is a scalable AI engine for impact-driven career alignment.

Climate Tech Career Explorer 🌍

AI-powered platform connecting students to climate tech opportunities. Built for TreeHacks 2026.

🎯 Overview

A comprehensive job discovery and career coaching platform with 4 integrated components:

Automated Aggregation - Scrapes climate tech jobs using Browserbase/Stagehand
Smart Categorization - AI-powered job categorization using Perplexity Sonar
Career Coach Chatbot - Personalized guidance using Fetch.ai uAgents
Resume Matcher - PDF parsing and job matching with Modal

🏗️ Architecture

climate-tech-career-explorer/
├── frontend/              # Next.js + React + TypeScript
├── backend/               # FastAPI orchestration layer
├── agents/                # Fetch.ai career advisor agent
├── components/
│   ├── 1-scraper/        # Browserbase job scraper
│   ├── 2-categorizer/    # Perplexity categorization
│   └── 4-resume-parser/  # Modal PDF parser
├── database/              # Supabase schema
├── shared/                # Shared resources (taxonomy)
└── docs/                  # Documentation

🚀 Quick Start

Prerequisites

Node.js 18+
Python 3.10+
npm/yarn
Git

1. Clone & Setup

# Clone repository
git clone <your-repo-url>
cd climate-tech-career-explorer

# Install root dependencies
npm install

# Install frontend & scraper dependencies
cd frontend && npm install
cd ../components/1-scraper && npm install
cd ../..

2. Setup Python Environments

# Backend
cd backend
python3 -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate
pip install -r requirements.txt
cd ..

# Agents
cd agents
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt
cd ..

# Categorizer
cd components/2-categorizer
pip install -r requirements.txt
cd ../..

# Resume Parser
cd components/4-resume-parser
pip install -r requirements.txt
cd ../..

3. Configure Environment

# Copy environment template
cp .env.example .env

# Edit .env with your API keys
# Required:
# - Supabase credentials
# - Browserbase API key
# - Perplexity API key
# - Fetch.ai agent config
# - Modal credentials

4. Setup Database

Create Supabase project at https://supabase.com
Run the schema:

# Copy schema contents from database/schema.sql
# Run in Supabase SQL Editor

Update .env with Supabase credentials

5. Run Development Servers

Option A: Run all services concurrently

npm run dev

Option B: Run services individually

# Terminal 1: Backend API
cd backend
source venv/bin/activate
uvicorn main:app --reload --port 8000

# Terminal 2: Frontend
cd frontend
npm run dev

# Terminal 3: Agents (HTTP Server)
cd agents
source venv/bin/activate
python http_server.py

# Terminal 4: Scraper (when ready)
cd components/1-scraper
npm run scrape

# Terminal 5: Categorizer (when ready)
cd components/2-categorizer
uvicorn main:app --reload --port 3002

📚 Documentation

API Endpoints

Backend API (Port 8000)

GET /api/v1/jobs - List jobs with filters
GET /api/v1/jobs/{id} - Get job details
GET /api/v1/categories - Get climate tech taxonomy
POST /api/v1/resume/upload - Upload resume
POST /api/v1/chat/session - Create chat session
POST /api/v1/chat/message - Send message to bot

Full API docs: http://localhost:8000/docs

Components

Component 1: Scraper

cd components/1-scraper
npm run scrape  # Run scraper

Scrapes jobs from ClimateBase and other sources, uploads to backend.

Component 2: Categorizer

cd components/2-categorizer
uvicorn main:app --reload --port 3002

Categorizes jobs using Perplexity Sonar API and shared taxonomy.

Component 3: Chatbot Agent

cd agents
source venv/bin/activate
python http_server.py

Fetch.ai agent HTTP server providing conversational career guidance.

Component 4: Resume Parser

cd components/4-resume-parser
modal deploy modal_app.py

Deploys Modal function for PDF parsing and job matching.

🗂️ Database Schema

Tables:

jobs - Job listings with categories
resumes - Uploaded resumes and parsed data
categories - Climate tech taxonomy
chat_sessions - Chat session tracking
chat_messages - Conversation history

See database/schema.sql for complete schema.

🌐 Frontend Pages

/ - Landing page with stats
/jobs - Browse jobs with filters
/chat - Career coach chatbot
/upload - Resume upload

🔧 Tech Stack

Frontend:

Next.js 14 (App Router)
React 18
TypeScript
Tailwind CSS
Supabase Client

Backend:

FastAPI
Python 3.10+
Supabase (PostgreSQL)
httpx (async HTTP)

AI/ML:

Fetch.ai uAgents
Perplexity Sonar API
OpenAI (optional)

Infrastructure:

Browserbase (scraping)
Modal (serverless functions)
Supabase (database + storage)

👥 Team Workflow

Branch Strategy

main                    # Production
├── dev                # Integration branch
├── feature/frontend   # Your work
├── feature/scraper    # Teammate 1
├── feature/categorizer # Teammate 2
└── feature/resume     # Teammate 3

Development Workflow

Create feature branch from dev
Work on your component
Test locally
Push and create PR to dev
Team lead merges after review

Testing Integration

# Test backend
curl http://localhost:8000/health

# Test frontend API client
cd frontend && npm run dev

# Test scraper -> backend flow
cd components/1-scraper && npm run scrape

# Test chatbot
cd agents && source venv/bin/activate && python http_server.py

📊 Climate Tech Taxonomy

8 main categories from Climate Tech Map:

Energy - Solar, Wind, Storage, Nuclear
Mobility - EVs, Public Transit, Aviation
Built Environment - Green Buildings, HVAC
Food & Agriculture - Alt Proteins, AgTech
Carbon Management - Capture, Credits, MRV
Circular Economy - Recycling, Materials
Climate Finance - Impact Investing, VC
Nature-Based - Forestry, Ocean, Restoration

See shared/climate-tech-taxonomy.json for full taxonomy with subcategories and keywords.

🎯 Demo Checklist

🐛 Troubleshooting

Backend won't start:

Check .env has correct Supabase credentials
Verify venv is activated
Check port 8000 is available

Frontend errors:

Ensure backend is running on port 8000
Check NEXT_PUBLIC_BACKEND_URL in .env.local
Clear .next folder and rebuild

Database errors:

Verify schema.sql ran successfully
Check Supabase project is active
Confirm service key has correct permissions

Agent not responding:

Check agent port 8001 is available
Verify Fetch.ai dependencies installed
Check agent logs for errors

📝 License

MIT License - Built for TreeHacks 2026

🙏 Acknowledgments

TreeHacks 2026
Climate Tech Map
Sponsor APIs: Browserbase, Perplexity, Fetch.ai, Modal, Supabase

Built with 💚 for climate action

Analysis

Compare with all teams

View

Metric

Figures cover GitHub contributors during the hackathon window. A co-authored commit counts in full for each author, so per-member totals add up to more than the whole-team figures.

Technology

Found in codeClaimed only

CSSIn code
FastAPIIn code
JavaScriptIn code
LangChainIn code
Next.jsIn code
OpenAIIn code
PythonIn code
ReactIn code
SQLIn code
StreamlitIn code
SupabaseIn code
Tailwind CSSIn code
TypeScriptIn code

13 of 13 appear in the indexed code.

AI coding agents

Claude CodeConfig · Commits

Detected from committed agent config files and commit authorship. Absence of a signal is not proof an agent was unused.

Codebase size

Source size

742 KB

Source files

Counts recognized source files only; vendored directories, binaries and lockfiles are excluded, so this is smaller than the repository on disk.

Repository

sahajpatel008/C3

139 files · 5.6 MB · @ 5268d6f

Structure

Interface
58 files · 42%
Screens, components and styles rendered to the user.
Application logic
25 files · 18%
Domain rules, services and shared utilities.
+5 more
Data & schema
15 files · 11%
Schema definitions, migrations and data access.

Supporting

Layers are inferred from where files sit in the tree, not from reading the code. A project that names its directories unconventionally will read oddly here — open the file browser to check anything the diagram implies.

Languages

Python43%
SQL38%
Markdown13%
TypeScript6%
CSS0%
JavaScript0%

Share of indexed source by file size. Binary and vendored files are excluded.

Dependencies

backend/requirements.txt

pypi · 25

email-validator
fastapi
gotrue
httpx
lancedb
langchain
langchain-community
langchain-core
langchain-openai
openai
pdfplumber
postgrest
pydantic
pydantic-settings
python-docx
python-dotenv
python-multipart
realtime
+7 more

components/2-categorizer/requirements.txt

pypi · 15

beautifulsoup4
lancedb
langchain
langchain-community
langchain-openai
openai
openpyxl
pdfplumber
plotly
PyPDF2
python-docx
python-jobspy
requests
sentence-transformers
streamlit

frontend/package.json

npm · 14

@supabase/supabase-js
axios
next
react
react-dom
+9 more

agents/requirements.txt

pypi · 8

fastapi
httpx
openai
pydantic
python-dotenv
uagents
uvicorn
websockets

components/1-scraper/requirements.txt

pypi · 7

certifi
charset-normalizer
idna
python-dotenv
requests
supabase
urllib3

components/4-resume-parser/requirements.txt