From Physics to Applied AI & Biotech: A Structured AI Transition Case

Student: Sean Joe Lee
University: National University of Singapore
Positioning Theme: From Physics to Applied AI & Biotech
Case Type: Structured AI Transition

Not a Computer Science Student

When Sean entered the AI+X program in April 2025, he was not a “pure CS student.”

His background was in Physics at NUS.

He wasn’t trying to deepen an existing AI identity.
He was exploring whether AI could become one.

That distinction matters.

Sean represents a growing category of transition talent — students with strong quantitative foundations but without formal computer science training, looking to build a legitimate path into applied AI.

2025: Building Breadth Before Specialization

In one year, Sean completed four AI+X PBLs:

• Machine Learning in Quantitative Finance – J.P. Morgan

• AI in Visual Computing

• AI in Natural Language Processing – Hugging Face

• AI & Computer Vision in Biotech – Novo Nordisk

Signal:

• Cross-domain AI exposure (Finance, CV, NLP, Biotech)

• Technical breadth before narrowing

• Execution stamina across multiple cycles

He wasn’t chasing one trend.
He was stress-testing his fit across domains.

When Simpler Models Win

One of Sean’s most notable outcomes:

“Why Simple Augmentation Outperforms CycleGAN for Pneumonia Detection”

https://program.blendedlearn.org/best-outcome/when-complexity-fails%3A-why-simple-augmentation-outperforms-cyclegan-for-pneumonia-detection

This project demonstrated something deeper than technical execution.

It showed:

• Model robustness thinking

• Experimental comparison discipline

• Understanding when complexity is unnecessary

He didn’t default to advanced architectures.
He evaluated trade-offs.

That’s research maturity.

Technical Depth Signals

Visual Computing – Deepfake Detection

• Adversarial robustness testing

• Domain shift evaluation

Amazon Operations Strategy (On-Site)

• Multi-criteria decision-making (MCDM) modeling

• Supply chain disruption simulation

• Systems-level framing

He received explicit praise for his holistic analytical structure — not just coding ability.

The Inflection Point: AI + Biotech

Across projects — especially in computer vision and biotech — Sean’s direction crystallized.

He began focusing on:

• Machine learning research

• Computer vision applications

• Biotech and medical AI

This is where exploration turned into conviction.

The transition wasn’t accidental.
It was structured.

Winter 2026: On-Campus Activation

Sean joined the AI+X Winter On-Campus Experience in Boston.

He worked on:

• Amazon Operations (on-site)

• AI in Visual Computing (on-site)

But the real shift happened outside the classroom.

He met researchers affiliated with:

• Broad Institute

• Harvard University

Many of them had transitioned from non-CS backgrounds.

For the first time, Sean saw that his path was not unusual.

It was viable.

Post-Program Momentum

After returning:

• He independently followed up with Broad researchers

• Scheduled deeper discussions

• Submitted a research application

• Began considering a potential academic move to Boston

He described the experience as “transformative.”

More than skills.

It changed how he saw himself.

Founder-Level Match

Strategically, we matched Sean with:

• Founder of a hardware innovation lab

• Recently completed Series A

• Focused on Computer Vision + Biotech

Now:

• They are in direct contact

• Conversations ongoing

• Exploring a potential 2026 internship

This is not hypothetical networking.

It is a live pipeline.

Program → Domain Clarity → Founder Connection → Internship Possibility.

Public Credibility Moment

Sean presented his generative medical model work at:

AI+X Meetup @ Kendall Square (Winter 2026)

Audience included:

• Founders

• Harvard researchers

• Industry practitioners

He later followed up independently with a Harvard researcher for coffee.

That’s ecosystem activation.

What This Case Represents

Sean’s journey is not about “learning AI.”

It represents:

• Non-CS → AI transition proof

• Longitudinal learning (4 online + 2 on-campus)

• Cross-domain AI capability

• Ecosystem access beyond classroom

• Founder-level relationship building

• Real internship pathway (2026)

A physics student entered exploring AI and exited with technical depth across ML domains, a focused interest in computer vision + biotech, active researcher relationships in Boston, and a live founder-level connection potentially leading to a 2026 internship.

That is AI identity transformation.