What are the responsibilities and job description for the Machine Learning Engineer Intern position at Putnam Science Academy Main Office?
AI Elite Basketball | Summer 2025
Duration: ~8 weeks (June 16 ~ Aug 11)
Onsite (Putnam, CT), with housing & meals supported through PSA’s campus
Type: Unpaid internship, with potential path into a deeper role beyond summer
Open to candidates with CPT, OPT, and H-1B
About Us
This internship is hosted by PAI (Precision Athletics Intelligence), a new AI athletic program by Putnam Science Academy (PSA) — one of the most elite basketball-focused high schools in the U.S.
- PSA has won 5 National Prep School Basketball Championships, in 8 years, most recently in March 2025
- The school’s mission is to deliver world-class private high school education while developing players for NCAA and NBA levels
- PAI is built to bring AI into high-performance sports and education, starting with this summer MVP project
This is PAI’s first technical initiative, aiming to create a foundational performance analysis platform for PSA’s nationally ranked basketball program — with high visibility and real-world application from day one.
What You’ll Build
You’ll join a small, focused team building an end-to-end system to:
- Automatically analyze practice footage
- Detect key actions (shooting, movement, defensive effort)
- Deliver structured feedback to coaches and players within minutes
This product is aiming China Market and it will serve elite athletes and coaching staff immediately, with long-term potential to scale across teams and domains.
Your Role
As an ML intern, you'll work on the core computer vision pipelines that power the system.
Responsibilities
- Use or fine-tune models like YOLOv8, OpenPose, or MediaPipe
- Build pipelines to extract training insights from video
- Process raw frames into structured data (e.g. player tracking, shot detection)
- Evaluate models on accuracy, reliability, and latency
- Deliver usable outputs via APIs to frontend/dev teams
- Write modular, reproducible code for experimentation and iteration
What We’re Looking For
Core Skills
- Strong Python skills; comfortable with Jupyter, scripting, and code structure
- Experience with PyTorch or TensorFlow — or fast learning capability
- Comfortable using OpenCV and working with image/video data
- Familiar with Git and collaborative development environments
- Fluent spoken Chinese (Mandarin)
Mindset
We’re looking for someone who:
- Has real confidence in their ability to learn fast and figure things out independently
- Can take vague or high-level product goals, and turn them into working code
- Works through ambiguity with speed, structure, and clarity
- Cares about doing real work that gets used — not just academic experiments
- Is genuinely interested in basketball and understands the game at a basic level
- Thrives in a builder-style environment with ownership, speed, and open problems
Bonus (Not Required)
- Projects involving video analysis, pose estimation, or CV pipelines
- Experience with DeepSORT, sports heatmaps, or action recognition
- Familiarity with serving models via FastAPI, Flask, or REST endpoints
- Background as a player, coach, or data analyst in sports
Who Can Apply
We welcome candidates from a variety of backgrounds:
- Undergraduates (junior/senior preferred) with strong project experience
- Master’s students in CS, AI, or related fields
- PhD students focused on applied machine learning
- Self-taught engineers — if you’ve built real things, we want to see them
We value your ability to build and think clearly over your academic label.
Why This Matters
This is not a typical early-stage internship.
While our tech team is just starting out, our platform isn’t. You’ll be building within a system that already has:
- A championship-level basketball program
- Immediate real-world users: athletes and coaches with daily training needs
- A founder with full access to decision-making, facilities, and execution
- A high-trust environment where things move fast, and feedback is real
In many ways, PSA provides what most startups seek after Y Combinator:
A live environment, institutional support, immediate demand, and the room to build and scale.
If you have:
- Strong learning ability
- Clear technical thinking
- Ambition to turn huge ideas into real systems
…and you're excited by sports, education, AI, and building things from scratch — you’ll thrive here.
