← Archive Get every issue free →

Gradient Descent

Sunday, April 19 2026

Good morning, Pierluigi — 10 items • ~3 min read

Today's Brief

Reduced inference cost curves are driving the rapid development of smaller, more efficient AI models, which is why multiple labs have released compact models this week. This shift is likely to have a second-order consequence on capital markets, where the increased adoption of AI in finance will lead to more sophisticated risk analysis tools, potentially altering the way you approach portfolio optimization. You should be considering how to integrate these new AI-powered risk analysis tools into your existing workflow, specifically by evaluating the potential impact on your portfolio's Sharpe ratio and adjusting your asset allocation strategy accordingly.

📊 AI & Data Tools
AI Agents Need Their Own Desk, and Git Worktrees Give Them One

Git worktrees can provide AI agents with their own 'desk', enabling parallel agentic coding sessions, which changes the way you approach collaborative coding and version control, specifically by implementing Git worktrees in your development workflow to improve efficiency.

 •  → Read

How to Learn Python for Data Science Fast in 2026 (Without Wasting Time)

Learning Python for data science can be accelerated by focusing on practical applications and projects, which changes the way you approach skill acquisition, specifically by prioritizing hands-on experience with Python libraries like Pandas and NumPy in your data analysis workflow.

 •  → Read

It Takes 2 Minutes to Hack the EU’s New Age-Verification App

The EU's new age-verification app can be hacked in just 2 minutes, highlighting the need for robust security measures, which changes the way you evaluate the security of AI-powered tools, specifically by conducting regular penetration testing and vulnerability assessments in your development pipeline.

 •  → Read

📈 AI in Finance
Crypto Hack Worth $290 Million Triggers DeFi Contagion Shock

A $290 million crypto hack has triggered a DeFi contagion shock, emphasizing the importance of robust security measures in AI-powered financial systems, which changes the way you assess risk in your investment portfolio, specifically by incorporating AI-driven risk analysis tools to identify potential vulnerabilities.

 •  → Read

Aevex CEO Speaks on Raising $320 Million in US IPO

Aevex has raised $320 million in a US IPO, demonstrating the growing interest in AI-powered finance, which changes the way you evaluate investment opportunities, specifically by considering the potential for AI-driven growth in your portfolio allocation strategy.

 •  → Read

CEO Jill Wyant Speaks on Madison Air IPO

Madison Air has completed a successful IPO, highlighting the potential for AI-powered companies to access capital markets, which changes the way you approach investment analysis, specifically by incorporating AI-driven metrics to evaluate company performance.

 •  → Read

🏃 AI in Sports
The biggest question facing every WNBA team in training camp

Every WNBA team faces unique challenges in training camp, emphasizing the need for data-driven analysis to inform coaching decisions, which changes the way you approach performance analysis, specifically by using AI-powered tools to identify key player metrics and optimize team strategy.

 •  → Read

Notre Dame spring football: A running diary of the Fighting Irish's open practice

Notre Dame's open practice has provided valuable insights into the team's performance, highlighting the potential for AI-powered analysis to enhance coaching decisions, which changes the way you evaluate player performance, specifically by incorporating AI-driven metrics to inform recruiting and player development strategies.

 •  → Read

Transfer rumors, news: PSG move for Arsenal star Martinelli

PSG is rumored to be interested in signing Arsenal star Gabriel Martinelli, demonstrating the complex dynamics of player transfers, which changes the way you approach talent acquisition and roster management, specifically by using AI-powered tools to analyze player value and optimize transfer decisions.

 •  → Read

🎧 Podcasts
Why AI Evaluation Science Can’t Keep Up (with Carina Prunkl)

AI evaluation science is struggling to keep pace with the rapid development of AI, highlighting the need for more effective evaluation methodologies, which changes the way you approach AI model selection, specifically by prioritizing transparency and explainability in your model evaluation workflow.

 •  → Listen

That's your edge for today.

See you tomorrow morning with the next gradient step.

Subscribe free → Share this issue
Gradient Descent • Powered by Groq • Sources: curated RSS across 15+ publications
Delivered to pierluigi.derogatis@live.com
SharePost on X