My hands-on experience with AI and machine learning started small but taught me big lessons. In my IT class at Stranahan High, I built a Python script that predicts basketball practice attendance using past data like weather, day of week, and game schedules. I used pandas to clean the data and scikit-learn to train a basic regression model. It only hit 71% accuracy, but the project showed me two things: data is messy, and models reflect the bias of whoever collects it. Our team is mostly commuters, so “distance from school” mattered more than I expected. That stuck with me. Outside class, I’ve been experimenting with Teachable Machine. I trained a model to identify JROTC uniform errors from photos — missing nametag, wrong gig line, scuffed shoes. It wasn’t perfect, but it helped new cadets self-check before inspection. I also used ChatGPT’s API with Python to make a study-bot for my Economics class. It quizzes us using our own notes and flags topics we miss repeatedly. Building it forced me to learn API calls, prompt design, and why AI fails when the input is vague. I don’t have an app in the App Store yet. What I do have is practice turning real problems into code. My criminal justice focus pushes me to ask: who does this AI help, and who could it harm? In JROTC we study leaders like Nehemiah. He had an assignment — rebuild the wall — and he made sure every person had a role. That’s how I see AI. It’s a tool, not a replacement. The hero’s job is to make sure the tool serves the whole community. So how do I foresee making an impact? I want to work where AI meets public trust. One area is policing and criminal justice. Departments use AI for predictive patrols, facial recognition, and risk assessment. But if the training data overrepresents certain zip codes, the model just rebuilds the same old wall with new bricks. I plan to major in criminal justice with a CS minor so I can audit those systems. I want to build or evaluate models that are transparent, explainable, and checked for bias before they affect someone’s freedom. I also care about access. At Stranahan, we have ESOL students who miss announcements because English isn’t their first language. I’m learning how speech-to-text and translation APIs work so I can prototype a tool that converts our morning announcements into text and audio in Creole and Spanish. Nehemiah didn’t ask if people deserved to build. He handed them bricks. Tech should do the same. I know I’m early in this field. I haven’t published papers or launched a startup. But I’ve debugged a model at 2am, I’ve seen bad data break good intentions, and I’ve led teams where trust matters more than rank. That’s my foundation. In five years, I want to be the person in the room who asks: does this AI include the people we forgot? Does it protect them? If the answer is no, we don’t ship it. If the answer is yes, we make sure everyone knows how to use it. Heroes of the Bible stayed committed to assignments that were bigger than themselves. My assignment is to make sure AI is a tool that builds walls up instead of putting them between people. That’s the impact I’m training for.