The catalog

300+ projects to build

Every project comes with a complete PRD, HLD & LLD. Filter by domain, difficulty or tech stack to find your next build.

30 projects

Sort

Easy

House Price Estimator With a Streamlit Front End and Flask API

Home buyers, sellers, and agents all anchor on price estimates, and even a modest predictive model becomes far more useful once it is reachable both by people and by other software. This project builds a price predictor with exactly those two doors into it. The intern trains a gradient boosting regression model with scikit-learn on housing data, then delivers it two ways: a Streamlit web app where users enter property details through form controls and receive an estimated price with supporting visuals, and a lightweight Flask REST API that returns predictions as JSON for programmatic callers. Building both surfaces from one model artifact teaches the intern to separate modeling from serving, and to keep a single source of truth for preprocessing so the app and the API can never disagree about a prediction. The intern finishes with practical experience in boosted tree models, model persistence, basic API design with Flask, and interface building with Streamlit — a compact tour of the skills that turn a regression notebook into something a small real estate business could genuinely put in front of customers.

real-estatedata-scienceflask+4

View project

Medium

Decision Tree Survival Modeling App With Interactive Pruning Controls

Decision trees are the most explainable models in machine learning, which makes them ideal for teaching — but only if learners can actually see the tree and watch it change. This project builds a web application around that idea using the classic Titanic passenger dataset, where every split has an interpretable story. The intern develops a Flask backend that trains a decision tree classifier with scikit-learn to predict passenger survival, served through a frontend built with HTML, CSS, and JavaScript. Users view a rendered visualization of the fitted tree, adjust pruning parameters such as maximum depth and minimum samples per split, and watch the tree restructure and its predictions change in real time — a direct demonstration of the overfitting-versus-generalization trade-off. Jupyter notebooks accompany the app to document data preparation and experimentation, and per-passenger predictions let users test scenarios interactively. The intern learns how tree construction, overfitting, and pruning interact — knowledge made durable by manipulating it directly — while gaining full-stack experience: serving a model with Flask, wiring a JavaScript frontend to a Python API, and packaging an educational tool others can learn from.

educationcssdata-science+7

View project

Medium

Document Clustering Web Service With Dendrograms and Linkage Options

Media teams that need to organize large text collections rarely want a notebook — they want a tool: upload documents, choose settings, download organized results. This project builds that tool around hierarchical clustering, the technique of choice when the number of natural groups is unknown in advance. The intern develops a Flask web application where users upload a set of text documents, which are vectorized and clustered with agglomerative methods from scikit-learn. The interface offers a choice of linkage strategies — ward, complete, and average — and renders dendrograms so users can inspect the merge structure and pick a cut point informed by the data rather than a guess. Cluster assignments are exportable as files, with pandas and NumPy managing the data handling, so the results feed directly into downstream editorial or archival workflows. The intern learns text feature engineering, the mechanics and trade-offs of linkage criteria, and dendrogram-driven cluster selection, while practicing web development with Flask — delivering an unsupervised learning capability as a finished, self-service product rather than an analysis that lives on one laptop.

mediadata-scienceflask+5

View project

Medium

Dockerized Survival Prediction Web App Using Decision Trees

Shipping a model inside a working, hosted web application is a distinct skill from training it, and this project practices that skill end to end on an approachable problem: predicting Titanic passenger survival with a decision tree. The intern builds a Flask application backed by a scikit-learn decision tree classifier, with pandas and NumPy handling the data preparation. Users upload passenger data through the browser, trigger model training, and review the results: a visualization of the fitted tree alongside evaluation metrics such as accuracy and a confusion matrix that show how well the model separates survivors from non-survivors. The application is then containerized with Docker, giving it a reproducible runtime that can be hosted anywhere a container runs — the step that turns a local demo into a deployable service. The project's arc — data upload, model training, evaluation display, containerized deployment — mirrors the lifecycle of real internal machine learning tools. The intern finishes able to speak concretely about serving models with Flask and packaging Python applications with Docker, with a hosting-ready artifact to show for it.

educationdata-sciencedocker+6

View project

Easy

Containerizing a Machine Learning App With Flask and Docker

The models that matter are the ones other people can run, and containerization is how modern teams guarantee that. This project is a focused first encounter with Docker in a machine learning context: take a working model, wrap it in a web service, and make it run identically on any machine. The intern trains a simple scikit-learn model in Python and serves it through a small Flask application with a prediction endpoint. They then write a Dockerfile from scratch — choosing a base image, installing dependencies, copying application code, and defining the startup command — and build and run the container locally, verifying that the service responds correctly from inside its isolated environment. Clear written instructions for building and running the image round out the project, because reproducibility includes the documentation, not just the artifact. Though deliberately small, the project delivers one of the most transferable skills in the field. The intern finishes understanding images versus containers, how dependencies get frozen into a build, and why containerized services behave consistently across machines — knowledge they will reuse in nearly every deployment task of their career.

educationdata-sciencedocker+3

View project

Medium

Property Listing Portal With Visit Scheduling and Media Tours

Buying or renting property usually means bouncing between listing sites, phone calls to schedule visits, and photos of unknown quality. This project consolidates the experience into one portal where properties are listed with rich media, visits are booked online, and agents manage their pipeline in a single place. The intern builds the portal on a Flask backend with MySQL, modeling properties, agents, users, and visit schedules, and handling image and video uploads so each listing can offer a genuine visual tour. Search and filtering let buyers narrow results by the criteria that matter, while authentication and distinct agent and user roles ensure listing management stays in the right hands: agents publish and update properties and confirm visit requests, and users browse, save, and schedule. The frontend combines Next.js, React, and TypeScript with Tailwind CSS for a clean, responsive browsing experience across devices. Delivering it teaches media-heavy backend design, relational search and filter queries, and role-aware interface construction, giving the intern a portfolio piece that mirrors how real estate marketplaces actually operate.

real-estateflaskfullstack+5

View project

Hard

Search-Driven Real Estate Listings and Market Analytics Platform

Real estate moves fast, and both agents and buyers need more than static listings: they need instant search across thousands of properties and analytics that reveal how markets are moving. This project builds a listings platform where search speed and data insight are the core product. The intern implements a Flask and Python backend with PostgreSQL as the system of record for properties, agents, buyers, and transactions, then integrates Elasticsearch so full-text and filtered property search returns results in milliseconds even at scale. JWT authentication and role-based access separate agents, buyers, and administrators; agents manage listings and upload property images, buyers search and save properties, and admins oversee the platform. Analytics views aggregate listing and pricing data into market trends. The frontend combines React and TypeScript with Tailwind CSS for a responsive, filter-rich search experience. The project teaches the intern to run a dual-store architecture, keeping a relational database and a search index in sync, a pattern used across serious marketplace products, and to layer authorization and analytics on top. It is a demanding, production-shaped build that shows mastery of fullstack engineering beyond simple CRUD.

businessreal-estateelasticsearchflask+7

View project

Hard

Pre-Sales Research and Competitor Analysis Automation Agent

Sales teams lose hours before every pitch to manual research on prospects, competitors, and market positioning, and the quality of that research varies with whoever had time to do it. This project builds an agentic AI system that does the legwork automatically and consistently. The intern develops agents with Langchain that plan and execute research tasks, using LlamaIndex to ingest and query documents and OpenAI models for analysis and synthesis. The system gathers prospect and competitor information, evaluates positioning, and produces strategy recommendations that sales teams in e-commerce and retail can act on directly. A Flask backend exposes the agent workflows as services, while a Streamlit interface lets users launch research runs, watch progress, and explore the resulting insights interactively, with the whole system implemented in Python. The project takes the intern beyond calling a language model into genuine agent engineering, including decomposing a business workflow into autonomous steps, grounding outputs in retrieved data, and packaging the result as a usable product. Completing it demonstrates practical, current AI engineering applied to a problem with clear commercial value.

businesse-commerceflasklangchain+4

View project

Hard

Stock Market Trend Analysis and Report Generation Agent

Financial analysts spend large parts of their week pulling stock data, spotting trends, and writing up findings, work that is essential, repetitive, and increasingly automatable. This project builds an agentic AI system that takes over the cycle from data retrieval to finished report. The intern constructs the system with Langchain orchestrating the workflow, LlamaIndex organizing and querying financial documents and data, and OpenAI models performing trend analysis and drafting narrative reports. Integrations with external market APIs supply real-time stock data, so analyses reflect current conditions rather than stale snapshots. A Flask backend, written in Python, exposes the agent's capabilities and manages scheduled and on-demand runs, feeding a customizable dashboard where financial analysts review generated trends, drill into the underlying data, and export reports. The project teaches the intern to combine live data feeds with retrieval-augmented reasoning and to make automated analysis trustworthy through grounding and transparency. Completing it demonstrates exactly the AI-plus-finance engineering profile that fintech firms and analytics teams are actively seeking.

businessfinanceflasklangchain+3

View project

Medium

Automated Pre-Sales Competitor Research Agent for Sales Teams

Before any serious sales call, account executives are expected to know the prospect's business, the competitors already courting them, and the angles most likely to land. In practice that research is done manually, minutes before the meeting, or skipped entirely, and deals are lost to better-prepared rivals. Sales and business development teams in retail and B2B settings need that preparation automated and standardized. The intern builds an agent that performs this pre-sales research on demand. A Flask application written in Python accepts a prospect or market segment as input, then uses LangChain to orchestrate a multi-step pipeline: gathering competitor and market data, extracting the facts that matter for a sales conversation, and calling OpenAI models to turn raw findings into a concise, actionable briefing with talking points and risk flags. The intern designs the chain structure, the prompts that keep outputs factual and sales-relevant, and the endpoints through which a sales team would request and retrieve briefings. By finishing the project the intern demonstrates agentic workflow design, LangChain chain composition, careful prompt engineering for business-critical output, and the ability to wrap an AI pipeline in a deployable Flask service that solves a measurable commercial problem.

businessretailflasklangchain+2

View project

Medium

Stock Trend Analysis and Report Generation Assistant for Investors

Retail investors and junior analysts spend evenings doing what institutional research desks do with entire teams: tracking price movements, reading company news, and assembling the picture into a view they can act on. The raw information is abundant; the time to structure it is not. A tool that automates trend analysis and drafts readable research reports would close much of that gap. The intern builds that tool as a Python web service on Flask. Market data and company documents are ingested and indexed with LlamaIndex so the system can retrieve relevant context on demand, and OpenAI models generate the analytical layer: trend summaries for selected tickers, explanations of notable movements grounded in the indexed material, and full report drafts combining the two. A user profile system personalizes the output, letting each user register their watchlist, risk tolerance, and preferred report depth so the same engine produces meaningfully different briefings for different investors. The intern implements the data ingestion, the retrieval and generation pipeline, and the profile-driven report customization. Completing the project demonstrates retrieval-augmented generation over financial data, document indexing with LlamaIndex, personalization architecture, and the judgment required to present AI-generated financial analysis responsibly.

businessfinanceflaskllamaindex+2

View project

Easy

Question Answering Starter App Built on GPT-4o

Every developer entering the AI field needs a first project that connects a real application to a frontier model, and doing it properly involves more than pasting an API call into a script: keys must be handled securely, errors and rate limits must be managed, and the round trip from user input to model response must be wired through a real web framework. That foundational plumbing is exactly what this project teaches. The intern builds a question answering web application in Python using Flask. The app presents a simple interface where a user types a question, the backend forwards it to OpenAI's GPT-4o with a well-structured prompt, and the answer is rendered back on the page. Around that core loop the intern implements the details that separate a demo from an application: environment-based API key management, graceful handling of timeouts and API errors, input validation, and a small amount of session context so follow-up questions feel coherent. The intern also experiments with system prompts to shape the assistant's tone and scope for an educational audience. Completing the project demonstrates command of the request-response lifecycle in Flask, correct and secure integration of a commercial LLM API, and prompt fundamentals, providing the base layer on which every more ambitious agentic project in the program is built.

educationagentic-aiflask+2

View project