Research Blog
How we build AI for clinical research — methods, results, and what we learned, written by the people who did the work.
Nat Biomed Eng | Making LLMs Reliable Data Science Copilots for Biomedical Research
How we built BioDSBench to evaluate LLMs on biomedical data science, found that analysis plans are key to success, and designed the DSWizard agent framework. From benchmark to platform to biomarker discovery.
Human–AI collaboration in medical literature mining: what role for proprietary data and domain workflow?
LEADS: a foundation model for human–AI collaboration in medical literature mining (Nature Communications, 2025). Why proprietary data and domain workflow matter for systematic review—and how LEADS performs.
Why Prototyping AI Agents Should Be Easier - Vibe Engineered 8 Agents in 1 day
Using our BioDSA agent framework, we reproduced eight biomedical AI agents in one day. The real bottleneck wasn't intelligence — it was architecture friction. Here's what we learned about designing for rapid agent prototyping.
Don't Delegate, Do It Yourself — The New Way to Work in the Age of AI
For fifty years, the best career advice was simple: learn to delegate. As you rise, stop doing and start managing. That advice is now wrong.
DeepEvidence: Empowering Biomedical Discovery with Deep Knowledge Graph Research
Generic Deep Research agents browse the web, but biomedical discovery requires navigating heterogeneous Knowledge Graphs. DeepEvidence is a specialized investigator that systematically explores biomedical KGs, achieving 40% accuracy on HLE-Medicine while frontier models hit only 3.3%.
The AI Regulation Revolution: How New FDA and EMA Guidance Opens the Clinical Trials Market
January 2025 marked a turning point for AI in clinical trials. The FDA and EMA released landmark guidance that doesn't just regulate AI — it legitimizes it as a core technology in drug development. Here's what it means for AI companies.
Training Agentic AI Searcher for Biomedical Literature: Supervised Fine-Tuning and Reinforcement Learning
Agentic AI systems like ChatGPT DeepResearch can browse the web, but biomedical literature is different. We care about high-quality evidence from published studies, clinical trials, and curated databases. Here's how we trained LLMs to become domain-specific literature search agents using SFT and RL.
Beyond the Prompt: The System Design Engineering Behind Production-Grade Agents
Anyone can build an AI agent in a weekend. But then 'Day Two' happens. Building a production-grade agent isn't an AI challenge; it's a distributed systems challenge. Here are five architectural pillars for building enterprise-ready agents.
Building AI for Drafting Clinical Trial Documents: RAG, Fine-tuning and Agentic AI Workflow
Generic AI models struggle with highly specialized domains like clinical trials—they don't know what to retrieve, improvise details that should never be improvised, and forget that regulators exist. Learn how we built AI systems for clinical trial document generation through retrieval, fine-tuning, and agentic workflows, from Trial2Vec to InformGen.
Adapting Large Language Models for Systematic Literature Review of Clinical Trials: Workflow versus AI Agents
Systematic literature reviews in medicine are more than just searching and summarizing papers—they're regulated processes requiring reproducibility and rigor. Learn how TrialMind-SLR turns PRISMA guidelines into an AI-powered workflow, and how the LEADS foundation model outperforms GPT-4o on clinical literature mining tasks despite being much smaller.
Understanding AI's New Power Duo: MCP and Agent Skills in TrialMind
An introduction to Model Context Protocol (MCP) and Agent Skills—two architectural patterns transforming isolated AI models into connected, knowledgeable agents. Learn how Keiji AI's TrialMind implements these concepts for clinical research tasks, enabling standardized data connectivity and modular domain expertise for pharmaceutical development.
Predicting Clinical Trial Success: From Machine Learning to Agentic AI and Beyond
Predicting whether a clinical trial will succeed sounds like science fiction. Discover how machine learning models like HINT and SPOT are finding patterns in past trials to forecast future outcomes, and how agentic AI systems are transforming clinical development.
Developing Generative AI for Clinical Trial Recruitment: From Patient to Trials
Patient recruitment is the headache no one talks about at parties. Trials stall. Budgets burn. Great ideas never meet the right people. Learn how TrialGPT uses generative AI to match patients with clinical trials.
Powering the Next Generation of Clinical Research: How MCP Unleashes TrialMind Agents
With pharmaceutical companies spending over $50 billion annually on clinical development while facing 90% failure rates, the strategic imperative for AI-driven innovation has never been clearer. Discover how MCP standardizes AI integration for clinical research.
Clinical Trials Demystified: An AI Researcher's Guide (Part 4) — AI Models in Clinical Trials
AI is increasingly being applied throughout clinical trial lifecycle. Discover how NLP, generative AI, computer vision, and predictive modeling are transforming trial design, patient recruitment, safety monitoring, and operational efficiency.
Clinical Trials Demystified: An AI Researcher's Guide (Part 3) — Clinical Trial Data & Standards
Clinical trials produce and rely on a variety of data and documents, all of which must adhere to certain standards to ensure integrity, interoperability, and compliance. Discover CDISC, MedDRA, ICH guidelines, and how standardization enables AI-driven innovation.
Clinical Trials Demystified: An AI Researcher's Guide (Part 2) — Clinical Trial Software & Vendors
Clinical development is supported by a suite of specialized software systems collectively referred to as 'eClinical' tools. Discover the landscape of EDC, CTMS, eTMF, and emerging AI-powered solutions transforming trial operations.
Clinical Trials Demystified: An AI Researcher’s Guide (Part 1) — Workflow for Clinical Trials
Clinical trials are essential to translating cutting-edge biomedical discoveries into lifesaving treatments. Discover how AI is revolutionizing this $80B+ industry and where your expertise can make transformative impacts.