NVIDIA BioNeMo Agent Toolkit Turns Biomolecular Models Into Callable Skills for AI Agents in Drug Discovery
AI scientists are becoming a new interface for scientific computing. These agents read papers, write code, generate hypotheses, call APIs, and inspect files. But science is not software engineering. No test suite turns green when a hypothesis is correct. Discovery stays iterative, uncertain, and grounded in the physical world. That gap is what NVIDIA is targeting. NVIDIA published a hands-on walkthrough for its BioNeMo Agent Toolkit. The argument is direct. A general coding agent pointed at
Get smart on it
NVIDIA published a toolkit that packages biomolecular models as documented, callable skills for AI agents. The toolkit addresses a gap in scientific computing: general coding agents cannot reliably produce new medicines without specialized tools, so these skills span protein folding, molecular docking, generative chemistry, genomics, and protein design. When agents had access to these skills, task completion rose from 57.1% to 100% and agents produced twice as many passing assertions per 1,000 tokens. The skills can run on hosted endpoints for quick access or be deployed locally for lower latency and repeated iteration.
Related news
OpenClaw Releases iOS and Android Companion Node Apps That Connect a Phone to a Self-Hosted AI Agent Gateway
OpenClaw just released native companion apps for iOS and Android. The iOS app is listed as ‘OpenClaw – AI that does things.’ Both apps are free to download. They are not standalone chatbots. Each phone becomes a node in a self-hosted agent network. The assistant itself runs on a separate Gateway. That separation is the whole design. TL;DR OpenClaw’s iOS and Android apps are companion nodes, not standalone assistants. The Gateway runs the agent; phones add came
PyGraphistry Implementation Workflow for Interactive Graph Intelligence Pipelines in Security Analytics and Risk Investigation
In this tutorial, we build an advanced, Colab-ready workflow around PyGraphistry for interactive graph analytics and visualization. We start by creating a realistic enterprise-style access dataset, transforming it into nodes and edges, and enriching the graph with risk scores, anomaly indicators, centrality metrics, community detection, and layout embeddings. We then use PyGraphistry to bind graph structure, visual encodings, labels, tooltips, and filtered subgraphs, and to generate local inter
