anishasingh-dev/openenv-ai-support-simulation-projects
0
1---2title: OpenEnv AI Support Simulation3emoji: 🤖4colorFrom: purple5colorTo: pink6sdk: docker7sdk_version: "0.0.1"8app_file: app.py9pinned: false10---11 12# 🧠 OpenEnv: AI Customer Support Simulation13141516 17## 📌 Overview18 19**OpenEnv** is a simulation environment for building and evaluating AI agents in a customer support setting.20 21The goal is to design an agent that can intelligently process incoming support tickets by:22 23* 🏷️ Classifying the issue category24* ⚡ Determining the priority level25* 🎯 Selecting the appropriate support action26 27The environment evaluates the agent using a **deterministic reward function**, making it ideal for experimentation, benchmarking, and learning.28 29---30 31## Tech Stack32- python33- AI agent logic34- Simulation environment35 36---37 38## 📂 Dataset39 40The dataset consists of **simulated customer support tickets** that mimic real-world scenarios, including:41 42* 💳 Billing issues43* 🛠️ Technical problems44* ❓ General queries45* 🤯 Messy / real-world user inputs (typos, urgency, informal tone)46 47Each ticket includes ground truth labels for:48 49* **Category**50* **Priority**51* **Action**52 53---54 55## 🧮 Reward Function56 57The agent is evaluated per ticket using the following scoring:58 59| Component | Score |60| --------- | ----- |61| Category | 0.4 |62| Priority | 0.3 |63| Action | 0.3 |64 65✅ **Maximum score per ticket = 1.0**66 67---68 69## ▶️ How to Run70 71### Run Baseline Agent72 73```bash74python baseline_agent.py75```76 77---78 79## 🐳 Docker Setup80 81### Build the Container82 83```bash84docker build -t support-env .85```86 87### Run the Container88 89```bash90docker run support-env91```92 93---94 95## 🗂️ Project Structure96 97```98.99├── dataset.py # Dataset definitions100├── env.py # Simulation environment101├── grader.py # Evaluation logic102├── baseline_agent.py # Simple rule-based agent103├── requirements.txt # Dependencies104├── Dockerfile # Container setup105└── README.md # Project documentation106```107 108---109 110## 📊 Example Output111 112```113Final Score: 20.10114```115 116---117 118## 🚀 Future Improvements119 120* Add ML/NLP-based intelligent agents121* Introduce noisy / adversarial inputs122* Expand dataset with more edge cases123* Add web-based visualization dashboard124 125---126 127## 🤝 Contributing128 129Contributions are welcome! Feel free to fork the repo, open issues, or submit pull requests.130 131