BonaVia Edu

BonaVia × ExpertEase | AI and Data Engineering Career Practicum

AI, Data Engineering & Machine Learning Program for High School Students

Guided by a Machine Learning Engineer at Meta, students complete an end-to-end AI and data project. They examine how raw data is cleaned, organised, transformed and moved through a pipeline before it reaches a machine learning model and informs a product decision. Using Python, data analysis and machine learning methods, students build a technical project and explain the relationship between the data, model, evaluation results and system design.

  • Grades 9–12
  • 12 Weeks
  • 120 Hours
  • Online
  • No Prior Python, Data or Machine Learning Experience Required
Apply for a Student Fit ReviewRead Student Stories
BONAVIA · CAREER PRACTICUMAI SystemAI SYSTEM | Understand Data · Design Pipelines · Train Models · Explain Results
01Data
02Model
03Product

Mentor & Program Supervisor

Pin（Campion）Qian

Machine Learning Engineer at Meta

  • M.S. in Artificial Intelligence Engineering, Carnegie Mellon University
  • Experience across AI systems, LLMs and data engineering
  • Work in ranking, recommendation, distributed data pipelines and multimodal models
  • Previous game-AI and reinforcement-learning research experience

Campion’s experience spans data infrastructure, machine learning models and product engineering. He will guide students in turning a broad AI idea into an executable problem, building a clear data workflow, selecting and evaluating models, and understanding how a technical result fits into a real product system.

What Makes This Program Different

Students do not simply call an existing model. They learn how an AI system works from data to product.

01

Guided by a Current Machine Learning Engineer

Students learn from a mentor with experience in large-scale data systems, ranking and recommendation, LLMs and product engineering, seeing how AI research becomes a working system.

02

One Data Chain Across the Full Project

Problem definition, data cleaning, feature design, model training, evaluation and the final presentation all build around the same project.

03

Data Engineering and Machine Learning Together

Students do not train a model in isolation. They examine where the data comes from, how it is processed and why data quality changes the result.

04

Students Explain More Than Accuracy

They must explain the selected data, assumptions, model performance and the limitations the system may face in real use.

From AI Curiosity to a Real Machine Learning System

A useful AI product does not begin with a model. It begins with a problem, data and judgment.

Students begin with a concrete problem and trace how data is collected, cleaned, organised, stored and transformed. They then examine how a model learns, how it is evaluated and how engineering teams turn a research result into a system that can support a product decision. Throughout the project, they ask one question: is this AI result reliable, explainable and appropriate for the problem it is meant to solve?

  1. 01Turn a broad AI idea into a clearly defined problem
  2. 02Understand raw data, quality concerns and boundaries of use
  3. 03Clean, aggregate, label and transform features
  4. 04Design a simplified data pipeline and experiment workflow
  5. 05Train and evaluate a machine learning model
  6. 06Explain model results, system limitations and product meaning

Three Connected Dimensions

From Data to Models to Real Products

The three dimensions form one end-to-end AI product journey: make the data reliable, make the model useful, then understand how the system enters a real product and continues to improve.

01

Data Engineering

Understand how raw information is collected, cleaned, organised and refreshed until it becomes reliable input for an AI system.

  • Identify data sources, fields and quality concerns
  • Clean, aggregate, label and extract features
  • Work with SQL, tables and simplified pipelines
  • Manage versions, refreshes and experiment records
02

Machine Learning

Use Python and machine-learning methods to build, train and evaluate a model, then explain whether the results support the project goal.

  • Understand training, validation and testing
  • Build models with Python and tools such as PyTorch
  • Compare performance and explain key metrics
  • Explore classification, ranking or recommendation tasks
03

Product and Systems Impact

Understand how a model enters a real product, how engineering teams manage iteration, and how technical choices affect the user experience.

  • Connect the model to a specific user and product problem
  • Understand deployment, inference and continued updates
  • Track experiments and compare alternatives
  • Judge trade-offs among accuracy, speed, cost and experience

The 12-Week Professional Journey

From raw data to an AI system that can be evaluated and explained.

Six stages build around one AI and data project. Students define the problem, establish the data foundation, train a model, understand the wider system and complete a final technical presentation.

  1. 01

    Weeks 1–2

    Problem Definition and the AI Data Lifecycle

    Choose a problem suited to the project scope, understand how data moves from creation to model input, and define the question the project must answer.

  2. 02

    Weeks 3–4

    Data Cleaning, Exploration and Quality

    Use Python, SQL or related tools to inspect the data, handle missing or unusual values, complete exploratory analysis and record limitations.

  3. 03

    Weeks 5–6

    Data Pipelines and Feature Engineering

    Design a simplified data workflow, complete aggregation, labelling, storage and feature extraction, and understand how the pipeline supports repeated experiments.

  4. 04

    Weeks 7–8

    Model Building and Evaluation

    Train and compare machine learning models, interpret metrics and errors, and test whether the results support the project question.

  5. 05

    Weeks 9–10

    AI System Design and Product Connection

    Connect data, model, inference and the user context into one system, considering experiment tracking, model updates and practical constraints. Students also design a focused AI tool prototype for a mentor-company data or model workflow and explain how data, models and human judgment work together.

  6. 06

    Weeks 11–12

    Final Technical Project and Presentation

    Complete the model, data workflow and system explanation, demonstrate the result, answer questions and identify next improvements.

Would you like to explore whether this learning path fits the student?

Apply for a Student Fit Review

What Students Complete

Every technical result must be connected to its data and explained clearly.

Together, these outcomes show how a student defines a problem, works with data, builds a model, evaluates evidence and translates technical work into product judgment.

01

AI Problem Definition and Project Scope

Explain the problem, intended use and what is deliberately outside the project scope.

02

Data Lifecycle and Field Map

Show how data moves from source through cleaning, storage, feature processing and model input.

03

Data Cleaning and Exploratory Analysis

Address quality concerns, identify patterns and explain limitations or bias within the data.

04

Simplified Data Pipeline

Create a repeatable data-processing workflow and document its main steps and dependencies.

05

Machine Learning Model and Evaluation

Build a model, choose metrics, compare performance and explain why the result is reasonable or still incomplete.

06

AI System Architecture

Explain the relationship between data, model, inference, updates and the intended product context.

07

Final Technical Presentation and Defence

Demonstrate the project, respond to questions and explain key technical choices, limits and next steps.

08

AI Data Tool Prototype for the Mentor Company

Design a working AI tool for a focused data-processing, model-evaluation or product-decision workflow.

Student Experiences

What Students Say

11 student perspectives

These reflections come from students across different BonaVia × ExpertEase career practicums. Each card identifies the student’s original program, mentor background and year.

01

Student J.X. | AI Career Program | Mentor background: Google | 2025

A Program That Completely Changed My Understanding of AI

“Before this program, AI mostly meant chatbots and homework tools to me. I did not realize how much more people were doing with it. Learning how major tech companies use AI was honestly mind-blowing. At school, a lot of students are still told not to use AI at all. Here, I learned how to use it to research, build, test ideas, and solve real problems while still thinking for myself. That completely changed the way I see AI. Most students my age have not had the chance to work with it this way yet, so I feel like I now have a really useful starting point. I would highly recommend the program to students who want to experiment with AI for themselves.”
02

Student A.L. | Software Engineering Career Program | Mentor background: Citadel | 2025

Gaining Technology and Industry Insight Beyond School

“My school had never given me this kind of exposure to technology or the industry. Through the apprenticeship, I started to see what the field looks like outside a classroom. Working with a mentor who is actively shaping the field made a huge difference. I was learning from someone with firsthand knowledge, which gave the experience a completely different weight. It changed what I thought was possible for my future and helped me see technology as more than something I studied at school. It became a path I could actually imagine for myself.”
03

Student D.H. | Private Equity & Investment Banking Career Program | Mentor background: Pender Software Holdings | 2025

From Not Understanding PE to Completing the Full Analysis Process

“Before this program, I honestly wasn’t sure what private equity was. Now I feel like I have a pretty solid idea of how it works. Going through a deal process was a big takeaway in itself because there is so much more to it than I expected. We worked through a CIM, wrote an investment memo, and had to decide whether a company was actually a good buy. That made private equity feel real instead of just being another finance term I had heard before. By the end, I could explain the process in my own words and understand why each step mattered. I learned a lot and would genuinely recommend this program to anyone who is curious about finance.”
04

Student P.S. | Private Equity & Investment Banking Career Program | Mentor background: Pender Software Holdings | 2025

Learning the Real Day-to-Day Work and Analytical Process

“The first things I learned were more of the day-to-day tasks, like building a CIM and writing an investment memo. We also talked about normalized EBITDA, competitive moats, debt, risk, and how to analyze whether a company is a good buy. At first, some of these terms sounded pretty confusing, but using them in a real project made them much easier to understand. This was not just memorizing definitions or making a basic presentation. We had to look at the information, question the assumptions, and explain our own decisions. I really liked that the work felt close to what people in finance actually do. It taught me much more than I could have learned from a textbook.”
05

Student F.S. | Private Equity & Investment Banking Career Program | Mentor background: Pender Software Holdings | 2025

Using Program Knowledge in Networking Conversations

“I didn’t know how impactful this program would be, but once I started going to networking events, the small points we learned made a really big difference. I remember reviewing our notes about debt and risk before one event and then using that knowledge to ask a much better question. The person I spoke with actually had a lot to say about it, and the conversation landed really well. Before the program, I would not have known what to ask or how to explain my question clearly. This is knowledge I probably never would have learned otherwise. The program made me much more confident speaking with professionals and helped me feel like I could actually take part in the conversation.”
06

Student A.W. | Finance Career Program | Mentor background: Pender Software Holdings | 2026

Seeing What Finance Careers Really Look Like and Discovering More Options

“Before the program, I didn’t really know what working in finance looked like. I honestly wondered, is it really just Excel spreadsheets every single day? Speaking with a mentor who actually works in the industry put the role into perspective and helped me understand what people really do. I also realized that finance is applicable to so many things, and that investment banking and consulting are not the only paths. The program opened me up to different options and possibilities in finance that I had never considered before. It gave me a much wider view of the industry and helped me think more clearly about what I might want to study or explore in the future. I would absolutely recommend it, especially if you are interested in finance but have no idea where to start.”
07

Student A.L. | Marketing Career Program | Mentor background: PwC | 2025

Creating a Launch Plan for a Real Brand

“I honestly thought market research was pretty straightforward: make a survey, find a few statistics, and put everything into a PowerPoint. Then we started working with a real local business, and I realized how much I had missed. We had to understand the customers, study the competition, plan the launch, and write a marketing plan the business could actually use. Some parts of our plan were carried out, which was probably the biggest surprise for me. It felt completely different from handing in an assignment and getting a grade. Something we created was useful in the real world. That was incredibly satisfying. I learned a lot.”
08

Student J.W. | Marketing Career Program | Mentor background: PwC | 2024

How a High School Experience Helped Me Land a Paid Internship

“I completed this program in Grade 12, and it became one of the most useful experiences on my resume. In my first year of university, it was a big reason I landed a paid internship. During the interview, I could explain the entire business process clearly, from market research and strategy to the launch and results. I could explain why we made each decision and answer follow-up questions because I had done the work in depth. I was not repeating marketing terms or trying to make a small school project sound bigger than it was. I knew what I was talking about. Getting that experience before I graduated from high school gave me a huge head start.”
09

Student A.L. | Biomedical Career Program | Mentor background: PhD mentor | 2026

Where Research Meets Business

“I joined because I was interested in biomedical science, so I assumed the program would focus mostly on research. I did not expect business to be such a big part of it. A great invention still has to become something people can understand, access, and use, and I had never really thought about that side before. The program taught me to consider the science behind an idea as well as what it takes to bring that idea into the real world. That connection between science and business became my favorite part. I am now planning my own product line, and this experience gave me a much clearer idea of where to begin.”
10

Student O.W. | Product Manager Career Program | Mentor background: P&G | 2024

Building a Mentor Connection and Getting an Early Career Head Start

“My mentor was awesome! She taught us, but what I remember most was how she connected with us and shared her own journey. That made the experience feel personal. I was learning from someone who was willing to talk about what her path had actually looked like. I also built connections through the program, and some of them have already opened doors to opportunities I never imagined. Getting that kind of exposure so early feels like a real head start on my career.”
11

Student A.H. | Graphic Design Career Program | Mentor background: Nike | 2023

Bringing Creative Ideas to Life in Real Projects

“It was amazing to see my ideas come to life in real projects. Working on real projects made the skills I was developing feel much more concrete. What stayed with me most was knowing that the work had a tangible impact. That was such a cool feeling. The experience also made my college applications stand out. By the time the program ended, I felt much more prepared for whatever comes next.”

01/11

Ready to explore whether the program is a good fit?

Apply for a Student Fit Review

Starting Point

Who Is This Program For?

  • Open to students in Grades 9–12
  • No prior Python, data science, machine learning or software-engineering experience required
  • Relevant to students exploring AI, LLMs, data analysis, software engineering, algorithms or technology products
  • Ready to invest time, work with data, test models and explain results

Personalized Learning Design

Each student begins from a different technical foundation and works toward an outcome that can be tested.

The team considers the student’s Python experience, mathematical understanding, data skills and pace of learning, then tailors preparation materials, foundational instruction and stage-specific support. A beginner can first establish programming, data and model foundations, while an experienced student can take on more complex pipeline design, model comparison and system architecture.

The project does not force every student through the same mechanical difficulty, but it does preserve the final standard. Each student must understand the data, model and judgment and explain independently how the project works.

01

Tailored Foundations

Preparation in Python, data analysis, mathematical intuition and machine learning is selected around the student’s starting point.

02

Personalized Project Depth

Data complexity, modelling methods and system-design challenges are adjusted within the same project framework.

03

Targeted Technical Feedback

Students receive stage-specific support around data quality, evaluation, code logic and technical communication.

Growth and University Preparation

The value is not in saying “I did AI.” It is in explaining the relationship between data, model and judgment.

The program title alone does not determine a university outcome. The value comes from whether the student personally worked with data, understood how a model was trained, compared results, handled uncertainty and can explain why the system is or is not suited to a problem.

These experiences can help students understand AI, data science, computer science and software engineering more realistically while giving them technical evidence and reflection they may later discuss in applications, résumés, portfolios and interviews.

Apply to the Program

Program Application

The program is open to students in Grades 9–12, with no prior Python, data or machine learning experience required. Preparation materials, technical pathways and project support are personalized to the student’s starting level so they can build the required foundation and progress through data processing, model development, system design and the final technical presentation.

Application Process

  1. 01Interest Form
  2. 02Fit Assessment & Interview
  3. 03Admission and Enrolment
Apply for a Student Fit Review

The regular format runs for 12 weeks, with an average commitment of 10+ hours per week. The summer intensive runs for about 1.5 months, with an average commitment of 20+ hours per week. The program is fully online.

Start Here

Complete the 1-Minute Student Fit Form

Tell us about the student’s current grade, AI and data interests and goals. BonaVia will consider the commitment, learning style and starting point, then help the family understand whether this experience is a good fit.

Selected programAI, Data Engineering & Machine Learning Program

1Family and student2Goals and contact
Step 1: Family and Student Information