🤖 OpenAI Wins Gold Medal at Competitive Coding Event

🥇OpenAI Wins Gold Medal at Competitive Coding Event

🔮Predicting a Person’s Risk of Developing a Disease

🛠Trending Tools

🥪Brief Bites

💰Funding Frontlines

💼Who’s Hiring?

The International Collegiate Programming Contest (ICPC) World Finals involves teams of three students from nearly 3,000 universities across 103 countries competing to solve complex algorithmic problems.

OpenAI developed an experimental LRM paired with GPT-5 that correctly solved 12 out of 12 of these exceptionally challenging algorithmic problems.

It reportedly relies on three main components:

If frontier AI models can solve exceptionally challenging algorithmic problems, it means they’re doing more than just performing sophisticated pattern matching; they’re demonstrating the ability to understand concepts, formulate strategies, and construct logical arguments.

It’s the first time in ICPC World Finals history that all the complex algorithmic problems were solved. This achievement also marks the first time a frontier AI model has displayed some form of superhuman intelligence within a specific domain.

It was trained on 400,000 anonymized clinical datasets from UK Biobank, which contained hospital visits with detailed timelines of diagnoses since birth.

Each medical event was classified using standardized medical codes, such as the International Classification of Diseases (ICD). For example, if a doctor diagnoses Type 2 Diabetes and prescribes Metformin to treat high blood sugar levels, that’s categorized as ICD Code E11.9.

“Delphi-2M” leveraged these standardized medical codes to convert complex medical records into structured sequences of health events, preserving the chronological order of hospital visits, diagnoses, and treatments to analyze how a patient’s health evolved over time.

As a result, it learned to simulate how diseases might progress throughout a patient’s lifespan based on demographics like age, gender, ethnicity, socioeconomic status, and geographic location.

It achieved an average AUC of approximately 0.85 over a 5-year period. In other words, “Delphi-2M” correctly distinguished between patients who’d develop a disease and those who wouldn’t 85% of the time.

It excels at predicting diseases with well-understood progression patterns such as hypertension, breast cancer, and osteoporosis. While promising, it struggles to perform well on rare diseases with irregular progressions.

Context: I’m planning for long-term financial security,

Clarity: but I’m not sure how to create a diversified investment portfolio that aligns with my {Insert Risk Tolerance} and {Insert Time Horizon}.

Guidance: Can you help me develop a long-term investment strategy based on {Insert Age}, {Insert Starting Capital}, and {Insert Monthly Contribution}?

Nothing raised a $200M Series C at a $1.3B valuation to pioneer AI-first consumer devices.

Irregular closed an $80M Series A at a $450M valuation to offer cutting-edge AI security.

Vibranium Labs landed a $4.6M Seed Round to fix bugs, alerts, and failures.

EarnIn (Mountain View, CA): AI Marketing Creative Intern, Winter 2025

NVIDIA (Austin, TX): ASIC Hardware Design Engineer, Entry-Level

Anthropic (Seattle, WA): Product Engineer, Applied AI, Mid-Level

Meta (Menlo Park, CA): AI-Focused Finance Manager, Senior-Level