NovaMentor – Autonomous Learning Agent

NovaMentor is an autonomous AI learning agent powered by Amazon Nova that turns student goals into smart study plans, summaries, revision schedules, and mock tests.

Comment

Inspiration

As students, we often set ambitious goals like “I will crack placements” or “I will master Machine Learning.” However, the real challenge is not motivation — it is structured planning. Most learners struggle to: Break large goals into smaller topics Allocate time effectively Design a proper revision strategy Measure learning progress Because of this, students spend more time planning what to study rather than actually studying. NovaMentor was built to solve this problem by acting as an intelligent AI study mentor that converts a high-level goal into a clear, structured learning roadmap.

What it does

NovaMentor is an AI-powered autonomous learning assistant built using Amazon Nova foundation models on AWS Bedrock. The user provides: . Study Goal . Number of Days . Difficulty Level Based on this input, the system generates:

  1. Structured Topic Breakdown
  2. Day-wise Study Plan
  3. Concept Summarie
  4. Practice Questions
  5. Revision Strategy The system models planning mathematically as: LearningPlan=f(Goal,Duration,Difficulty) Where: 𝐺𝑜𝑎𝑙 defines scope 𝐷𝑢𝑟𝑎𝑡𝑖𝑜𝑛 defines scheduling constraint 𝐷𝑖𝑓𝑓𝑖𝑐𝑢𝑙𝑡𝑦 controls conceptual depth This ensures adaptive structured reasoning instead of generic text generation. ## How we built it NovaMentor is developed as a Streamlit web application using Python.

System Architecture

Frontend → Streamlit Backend → Python AI Engine → Amazon Nova Cloud Layer → AWS Bedrock

Technical Flow

  1. The user enters inputs through a clean sidebar interface.
  2. The system validates the inputs.
  3. A structured prompt is generated dynamically.
  4. The prompt is sent to Amazon Nova using AWS Bedrock via boto3.
  5. The model processes the request and returns structured output.
  6. The output is displayed in organized sections. Technologies Used Python Streamlit Amazon Bedrock Amazon Nova Foundation Model Boto3 python-dotenv Logging The backend is modular and includes functions like:

build_prompt() get_bedrock_client() generate_study_plan()

This keeps the code clean and scalable.

Challenges we ran into

While building NovaMentor, we faced several challenges: . Correctly configuring Amazon Bedrock . Formatting JSON requests properly . Managing AWS credentials securely . Parsing structured AI responses . Refining prompts to ensure consistent output Getting well-structured responses required multiple improvements to the prompt design.

Accomplishments that we're proud of

. Successfully integrated Amazon Nova via AWS Bedrock . Built a fully functional AI-powered web application . Implemented secure environment configuration . Designed modular and reusable backend logic . Created a clean and professional user interface Most importantly, NovaMentor generates meaningful and structured study plans instead of generic AI responses.

What we learned

Through this project, we learned: . How to work with Amazon Bedrock and foundation models . Practical prompt engineering techniques . Building AI-powered web applications . Secure AWS configuration practices . Writing clean and modular Python code This project gave us real hands-on experience in integrating AI into a working system.

What's next for NovaMentor – Autonomous Learning Agent

In the future, we plan to: . Add progress tracking . Include performance analytics . Implement user authentication . Add adaptive learning recommendations . Integrate voice interaction using Nova Sonic Our long-term gol is to make NovaMentor a smart and reliable learning companion for students.

Built With

Share this project:

Updates