SmartTutor – AI Study Buddy

InspirationHere’s a concise “About the project” narrative, organized into four sections:

Inspiration As a former high school student, I saw friends struggling to find affordable, on-demand tutoring outside school hours. Watching classmates cram for exams with little feedback inspired me to build SmartTutor—an always-available, AI-driven study buddy that adapts to each learner’s pace.

What I Learned Building SmartTutor deepened my understanding of conversational AI pipelines. I gained hands-on experience defining intents and utterances in Amazon Lex, tuning natural-language responses, and orchestrating AWS Lambda functions to generate on-the-fly quizzes. Integrating AWS Transcribe and Polly taught me the nuances of speech-to-text accuracy and lifelike text-to-speech.

How I Built It I scaffolded a React front end and hosted it via AWS Amplify (Cognito for auth). I created two core Lex intents—“ExplainConcept” and “GenerateQuiz”—and wrote a Python Lambda that analyzes user text to produce multiple-choice questions. DynamoDB stores each student’s quiz history, while S3 houses static diagrams. PartyRock’s webview wraps the entire app, letting us deploy quickly to both browser and mobile.

Challenges Faced The biggest hurdle was quiz-generation logic: extracting key terms from arbitrary text and turning them into meaningful questions. I iterated between simple keyword extraction and fine-tuned NLP prompts to strike a balance between speed and question quality. Ensuring smooth voice interaction across accents required adjusting transcription settings and handling mis-recognitions gracefully in the UI.

What it does

How we built it

Challenges we ran into

Accomplishments that we're proud of

What we learned

What's next for SmartTutor – AI Study Buddy

Built With

• amazon-lex/polly/transcribe
• and
• aws-amplify-(cognito
• aws-sdk)
• built-with-javascript-(react
• deployed
• dynamodb
• hosting)
• lambda-&-api-gateway
• python
• s3
• via