I lead the design and development of agentic AI applications at Alloy Therapeutics — tools that help scientists explore experimental, proprietary, and research data and turn it into actionable scientific insights.

I love owning things end to end: from a first idea with a scientist, to a hardened, monitored, production-grade AI product people actually use every day.

• Agentic AI: designing and building AI agents on modern LLM platforms and agentic architectures
• Model customization: evaluating fine-tuning and optimization strategies with Python and PyTorch
• Trustworthy AI: evaluation frameworks, monitoring, security controls, and governance for regulated scientific environments
• Cloud & data: integrating AI services with AWS infrastructure and enterprise data platforms
• Engineering discipline: CI/CD, automated testing, and deployments that scale

• Side by side with scientists, bioinformaticians, product teams, and engineers to find the high-impact AI opportunities
• Mentoring engineers and helping shape our long-term AI platform strategy
• The goal: next-generation AI assistants that speed up scientific discovery and make everyone’s day a little easier

LLMs · Agentic AI · Python · PyTorch · PostgreSQL · vector, relational & non-relational databases · AWS · CI/CD · Cloud Architecture · AI Evaluation & Monitoring

• LinkedIn: https://www.linkedin.com/in/gianpaolosantopaolo/
• Personal GitHub: https://github.com/gsantopaolo
• Blog: https://genmind.ch/