AI Engineering Internship Assignment: Model Deployment
with FastAPI
Topic Chosen: Deploying Machine Learning Models with
FastAPI
This assignment will test your ability to:
• Learn a technical AI engineering concept (model deployment).
• Teach it back to your peers using a clear, well-structured presentation.
• Demonstrate deep technical understanding through a full code walkthrough.
• Solve a real-world challenge and communicate your solution effectively.
1. Learning Resources
Use the following curated references to master the topic:
1. O%icial Documentation
• FastAPI Official Docs:
https://fastapi.tiangolo.com
The authoritative guide for FastAPI, including tutorials, deployment, and advanced
features1.
• FastAPI Docs (DevDocs mirror):
https://devdocs.io/fastapi/
Fast, searchable documentation for FastAPI2.
2. Beginner-Friendly Tutorials
• DataCamp FastAPI Tutorial:
https://www.datacamp.com/tutorial/introduction-fastapi-tutorial
Step-by-step introduction to FastAPI for building APIs, with plenty of code examples3.
• GeeksforGeeks FastAPI Introduction:
https://www.geeksforgeeks.org/python/fastapi-introduction/
Covers FastAPI basics, setup, and a simple example API4.
• FastAPI ReadTheDocs:
https://fastapi-tutorial.readthedocs.io
Community-maintained FastAPI documentation and guides5.
3. Model Deployment with FastAPI (Blogs & Guides)
• Deploying ML Models with FastAPI and Docker (Dev.to):
https://dev.to/code_jedi/machine-learning-model-deployment-with-fastapi-and-docker-llo
Detailed walkthrough for deploying a scikit-learn model as an API, including
Dockerization6.
• Founding Minds: FastAPI & Docker for ML Deployment:
https://www.foundingminds.com/deploying-ml-models-with-fastapi-and-docker/
Explains deployment best practices, AWS integration, and production tips7.
• GeeksforGeeks: Deploying ML Models as API using FastAPI:
https://www.geeksforgeeks.org/machine-learning/deploying-ml-models-as-api-using-
fastapi/
Step-by-step guide for deploying a model as an API, with code and explanations8.
• TestDriven.io: Deploying and Hosting a ML Model with FastAPI:
https://testdriven.io/blog/fastapi-machine-learning/
Covers deploying a stock prediction model with FastAPI and Heroku, including Docker
and CI/CD9.
• Uptrace: How to Use FastAPI [Detailed Python Guide]:
https://uptrace.dev/blog/python-fastapi
In-depth FastAPI guide with advanced features and deployment advice10.
4. Video Tutorials
• Deploy ML Models as APIs with FastAPI (Full YouTube Tutorial):
https://www.youtube.com/watch?v=0sOvCWFmrtA
Comprehensive, hands-on video for deploying ML models with FastAPI11.
• FastAPI Tutorial: Build a REST API in 15 Minutes:
https://www.youtube.com/watch?v=iWS9ogMPOI0
Quick, practical intro to building APIs with FastAPI12.
• Deploy ML Models with FastAPI, Docker, and Heroku:
https://www.youtube.com/watch?v=h5wLuVDr0oc
End-to-end deployment, including Dockerization and cloud deployment11.
• FastAPI Deployment Tutorials Playlist:
https://www.youtube.com/playlist?list=PLZoTAELRMXVPgsojPOHF9i0u2L83-m9P7
Multiple videos on API creation and ML model deployment13.
• How to Deploy ML Solutions with FastAPI, Docker, & AWS:
https://www.youtube.com/watch?v=pJ_nCklQ65w
Full-stack deployment, including cloud integration and best practices14.
5. Advanced/Production Deployment
• FastAPI in Containers (Docker):
https://fastapi.tiangolo.com/deployment/docker/
Official guide to containerizing FastAPI apps for production15.
• Northflank: Step-by-Step ML Model Deployment:
https://northflank.com/blog/how-to-deploy-machine-learning-models-step-by-step-guide-
to-ml-model-deployment-in-production
Covers containerization, CI/CD, and infrastructure setup16.
6. IDE/Development Environment
• VS Code FastAPI Tutorial:
https://code.visualstudio.com/docs/python/tutorial-fastapi
How to build, debug, and run FastAPI apps in Visual Studio Code17.
7. Real-World Use Case Example
• Deploying ML Models with FastAPI and Azure:
https://datastud.dev/posts/ml-fastapi/
Walks through deploying a model with FastAPI on Azure, including project structure and
CI/CD basics18.
2. Assignment Instructions
A. Learn & Summarize
• Study the above resources to understand:
o What is FastAPI and why is it used for model deployment?
o How do you build, test, and deploy an API for a machine learning model?
o Best practices for productionizing ML models.
B. Teach the Concept
• Prepare a PPT (8–12 slides) that:
o Explains FastAPI and its relevance in AI engineering.
o Illustrates the steps to deploy a machine learning model as an API.
o Uses diagrams, code snippets, and real-world analogies.
o Is tailored for new joiners with basic Python/ML background.
C. Code Walkthrough
• Choose a simple ML model (e.g., Iris classifier, sentiment analysis).
• Build and save the model in Python (using scikit-learn or similar).
• Develop a FastAPI application to serve predictions.
• Full Code Walkthrough:
o Explain each section: model loading, API endpoints, request/response structure,
error handling, and testing.
o Discuss the math behind the model (e.g., how logistic regression or decision trees
work).
o Highlight deployment considerations (e.g., Docker, scaling, monitoring).
D. Real-World Challenge
Scenario:
A client wants to automate flower species identification from petal/sepal measurements (Iris
dataset). They need a REST API that receives measurements and returns the predicted species.
• Build the solution end-to-end:
o Train a model on the Iris dataset.
o Deploy it as a FastAPI service.
o Demonstrate the API with sample requests.
E. Video Submission
• Record a video [NO time constraint] :
o Teach the FastAPI deployment concept using your PPT.
o Walk through your code in detail, explaining logic, math, and design choices.
o Demo your API solving the real-world challenge.
• Upload the video to Google Drive (or similar) and share a link with download or read
access so that we could evaluate it.
3. Submission Checklist in the drive
• PPT file (Google Slides or PDF)
• Well-commented code (GitHub or zip)
• Video link (Google Drive with download/read access)
• Short summary (2–3 sentences) of your project and learnings [.docx]
4. Evaluation Criteria
• Clarity and depth of concept explanation
• Quality and accuracy of code walkthrough (including math and technical reasoning)
• Effectiveness of teaching (PPT and video)
• Relevance and completeness of real-world solution
• Communication skills and ability to simplify technical content
Tip:
Teaching a concept simply is the best test of your own understanding. Focus on clarity, structure,
and real-world relevance.
