Job Application Assistant

A Job assistant built with CopilotKit (Next.js) on the frontend and DeepAgents (by LangChain) on the backend. Users upload their resume (PDF), the system extracts skills and context, and DeepAgents orchestrate sub-agents & tools to search the web (via Tavily) for relevant job postings. Results stream back to the UI in real time and are rendered alongside the chat.

DeepAgents provides clean orchestration with sub-agents and tools, while CopilotKit (AG-UI) handles real-time streaming and stateful UI updates.

What This Demo Shows

• Resume upload + PDF parsing
• Skill extraction from real resumes
• DeepAgents orchestration with sub-agents and tools
• Internet search via Tavily
• Tool calls streamed to the UI using AG-UI

High-Level Flow

[User uploads resume & submits job query]
        ↓
Next.js UI (ResumeUpload + CopilotChat)
        ↓
useCopilotReadable syncs resume + preferences
        ↓
POST /api/copilotkit (AG-UI protocol)
        ↓
FastAPI + DeepAgents (/copilotkit endpoint)
        ↓
Resume context + skills injected into agent
        ↓
DeepAgents orchestration
   ├─ internet_search (Tavily)
   ├─ job filtering & normalization
   └─ update_jobs_list (tool call)
        ↓
AG-UI streaming (SSE)
        ↓
CopilotKit runtime receives tool result
        ↓
Frontend captures tool output
        ↓
Jobs rendered in table + chat stays clean

Project Structure

.
├── src/                               ← Next.js frontend
│   ├── app/
│   │   ├── page.tsx
│   │   ├── layout.tsx                 ← CopilotKit provider
│   │   └── api/
│   │       ├── upload-resume/route.ts ← upload endpoint
│   │       └── copilotkit/route.ts    ← CopilotKit AG-UI runtime
│   ├── components/
│   │   ├── ChatPanel.tsx              ← Chat + tool capture
│   │   ├── ResumeUpload.tsx           ← PDF upload UI
│   │   ├── JobsResults.tsx            ← Jobs table renderer
│   │   └── LivePreviewPanel.tsx
│   └── lib/
│       └── types.ts                   ← Shared frontend types
│
├── agent/                             ← DeepAgents backend
│   ├── main.py                        ← FastAPI + AG-UI endpoint
│   ├── agent.py                       ← DeepAgents graph & tools
│   ├── pyproject.toml                 ← Python deps (uv)
│   └── .env.example
│
├── package.json
├── next.config.ts
└── README.md

Environment Variables

You will need a Google API Key and Tavily API Key.

Create agent/.env and set your keys:

GOOGLE_API_KEY=your-google-api-key
TAVILY_API_KEY=tvly-dev-...
GEMINI_MODEL=gemini-2.0-flash

Setup & Installation

1. Install Dependencies

Frontend (Next.js):

npm install
# or
yarn install

Backend (Python, uv):

cd agent
uv sync

The backend uses uv for dependency management. Install it if needed:

pip install uv

2. Running Locally

Start the backend:

cd agent
uv run python main.py

Backend runs on http://localhost:8123.

Start the frontend (in a new terminal):

npm run dev
# or
yarn dev

Navigate to http://localhost:3000 in your browser.

Usage

1. Open the app at http://localhost:3000
2. Fill in your target job title and location preferences
3. Upload your resume (PDF or TXT)
4. Chat with the assistant: e.g., "Find frontend engineer jobs in Remote or Bangalore with React/Next.js"
5. Watch as the agent searches the web and displays matching jobs in real-time

Tech Stack

Frontend:

• Next.js 16 (App Router)
• React 19
• CopilotKit
• Tailwind CSS 4
• TypeScript

Backend:

• Python 3.12+
• FastAPI
• LangChain DeepAgents
• Google Gemini
• Tavily Search API

License

This project is licensed under the MIT License.