Inspiration
Eleven years ago, I was diagnosed with Stage IV kidney cancer. As an engineer, I kept asking one question: Could AI transform the overwhelming volume of biomedical research into scientific intelligence? That question became the foundation of KCIP.
What it does
KCIP is an AI-powered scientific intelligence platform that starts with a scientific question, orchestrates specialized AI agents, analyzes biomedical evidence, and generates structured, traceable Executive Scientific Reports based on available evidence.
How we built it
KCIP was built using GPT-5.6 for scientific reasoning, multi-agent orchestration, evidence synthesis, and report generation. Codex accelerated software architecture, implementation, and iterative development. The platform was designed around governance, traceability, and reproducible validation.
Challenges we ran into
The biggest challenge was not building AI agents—it was designing a system that remains scientifically rigorous. We had to balance AI flexibility with evidence evaluation, governance, reproducibility, and transparent reasoning.
Accomplishments that we're proud of
We're proud of building a platform that prioritizes evidence over speculation. KCIP successfully completed its first controlled validation, demonstrating that responsible AI can acknowledge when current scientific evidence does not support a validated answer.
What we learned
We learned that the best AI systems do not replace scientific judgment—they augment it. Clear governance, traceable reasoning, and human oversight are as important as powerful AI models.
What's next for KCIP — Kidney Cancer Intelligence Project
Our next goal is to expand KCIP beyond kidney cancer into additional medical conditions and scientific research domains while continuing to improve its multi-agent architecture, validation framework, and evidence-based reasoning.
Log in or sign up for Devpost to join the conversation.