Data Scientist

Lightning launches new suite of tools for PyTorch Developers and researchers. Lightning, creators of PyTorch Lightning, today announced a suite of new tools built to accelerate distributed training, reinforcement learning, and experimentation for PyTorch developers and researchers. This press release features multimedia. View the full release here: https://www.businesswire.com/news/home/20251023426558/en/ Lightning's AI code editor where AI helps debug, train, inference & ship PyTorch on GPUs. The launch comes as the PyTorch community gathers for PyTorch Conference 2025, reflecting Lightning's continued commitment to building the best platform for PyTorch developers and researchers. The new suite introduces a new AI Code Editor purpose-built for PyTorch development inside Lightning Studios, Lightning Environments, a hub for interactive and large-scale training, an official integration with Meta's Monarch, making large-scale distributed training interactive, and zero-day support for OpenEnv, a new open standard from Meta for reinforcement learning. AI Code Editor: PyTorch Expertise, Built In The new AI Code Editor brings specialized, domain-aware assistance directly into Lightning Studios and Notebooks. Developers can tap into PyTorch-focused "experts" for training, inference, or reinforcement learning tasks. These experts help them build, debug, optimize, and deploy code faster inside a single cloud-native environment. They can also leverage Lightning's Model APIs to access any AI model, open or proprietary, directly from the same workflow with built-in usage tracking, access control, and billing via Lightning credits. The AI code editor is natively integrated with Lightning's GPU Marketplace. Developers can provision the exact GPU resources they need directly from the editor, so experiments can be launched immediately without any manual cluster setup. This allows PyTorch developers to move from code to execution seamlessly, whether running on a single GPU for rapid iteration or scaling across multiple nodes for large-scale training. "Our goal is to make every developer in the world a PyTorch developer," added William Falcon, CEO of Lightning AI and creator of PyTorch Lightning. "Whether you're training a model on one GPU or hundreds, Lightning gives you the same tight, interactive development loop people love, now supercharged by agents and instantly connected to the compute you need." Get started with Lightning's AI code editor for free. Lightning Environments provide self-contained, interactive spaces where developers can explore, train, and scale reinforcement learning and distributed AI workloads. Each environment is like a house in a growing neighborhood. It is a ready-to-use setup that can scale from a single GPU to multi-node clusters with no infrastructure overhead. Developers can "unlock" one house, complete experiments, and then move to the next, building more complex multi-agent systems or progressive reinforcement learning experiments. Environments also serve as evaluations and sandboxes, letting developers safely test new models, agents, or workflows in a controlled, reproducible space. Developers can share their setups through the new Environments hub, Lightning's growing library of open, reproducible research workspaces. Monarch and Lightning: Empowering the Next Generation of AI Builders In collaboration with Meta's PyTorch team, Lightning is bringing Monarch directly into Lightning Studios, combining the power of large-scale distributed training with the ease, speed, and fully interactive experience of local notebook development. Monarch makes cluster-scale training interactive, persistent, and accessible. It allows developers to iterate on experiments, debug, and modify code in real time without restarting or re-allocating compute resources. Integrated with Lightning's Multi-Machine Training service, developers can scale from a single notebook to hundreds of GPUs across multiple cloud providers, all within familiar PyTorch workflows. "Monarch redefines what distributed training feels like," said Luca Antiga, CTO of Lightning. Antiga, who serves as Chair of the Technical Advisory Council of the PyTorch Foundation and authored Deep Learning with PyTorch, brings deep PyTorch expertise to this collaboration. "Together with Meta's PyTorch team, we're making large-scale development as interactive and flexible as local experimentation. This empowers the next generation of AI builders to move faster than ever." Lightning Enables Zero-Day Support for OpenEnv, Standardizing Reinforcement Learning Environments Lightning is announcing zero-day support for OpenEnv, a new open standard from Meta that defines how reinforcement learning environments are packaged, shared, and run. OpenEnv makes it easy for researchers and environment creators to develop rich, reproducible experiments. With Lightning, developers can run any OpenEnv environment locally or scale it seamlessly across multiple GPUs and distributed across clouds, all inside isolated, reproducible sandboxes. Lightning extends OpenEnv with enterprise-ready infrastructure, providing full GPU access, advanced networking, monitoring, and security controls for both research and production workloads. torchforge: Built on Monarch, Powered by Lightning Lightning also adds support for torchforge, Meta's new PyTorch-native framework for reinforcement learning, built on top of Monarch. torchforge provides a clean, composable interface for authoring RL algorithms while scaling seamlessly across clusters. Researchers can now run torchforge experiments natively on Lightning, taking full advantage of distributed OpenEnv environments for RLHF and other large-scale training workloads. Building the Home for PyTorch Developers With these launches, Lightning cements its position as the go-to platform for PyTorch developers, combining scalable compute, distributed training frameworks, and AI-assisted development in one unified experience. To explore the new Environments, AI Code Editor, and Meta integrations, visit lightning.ai. View source version on businesswire.com:https://www.businesswire.com/news/home/20251023426558/en/

Newsletter Signup - Under Article / In Page"*" indicates required fields Artificial intelligence (AI) has taken the biotech industry by storm, allowing companies to speed up the drug discovery process while also making it more cost-effective. With so many companies in the industry now embracing the technology, we take a look at 12 AI drug discovery companies. The COVID-19 pandemic revealed AI to be an essential tool in helping to find treatments and vaccines with greater speed and precision. Since then, there have been several drug discovery breakthroughs for AI within the biopharma industry, from helping to quickly and efficiently discover a new antibiotic called abaucin to combat a multi-drug resistant bacteria, to fully discovering and designing a drug that has entered clinical trials. Here are 12 AI drug discovery companies currently making great strides with their technology. Anima Biotech Technology: mRNA biology modulators Disease areas: Immunology, oncology and neuroscienceRecent news: Announced promising preclinical data for lead pulmonary fibrosis candidate Anima Biotech’s AI drug discovery technology is built around its mRNA Lightning.AI platform, which images hundreds of cellular pathways in both healthy and diseased cells to train disease-specific AI models, making use of neural networks to help these models distinguish between healthy and diseased cells and identify dysregulated pathways. These pathways are subsequently analyzed to uncover novel targets backed by experimental validation. Anima currently has 20 preclinical candidates being evaluated for immunology, oncology, and neuroscience indications, with its most advanced candidate indicated for the treatment of lung fibrosis. The company announced in February 2024 that this candidate had shown promising preclinical results and could open up new avenues for treating patients with idiopathic pulmonary fibrosis. The AI drug discovery company also has ongoing collaborations with several pharma giants. After initially partnering with Eli Lilly in 2018 and Takeda in 2021, its most recent partnership was formed with AbbVie in 2023 for the discovery and development of mRNA biology modulators against oncology and immunology targets. Atomwise Technology: TYK2 inhibitor  Disease area: Autoimmune and autoinflammatory diseasesRecent news: Published results showcasing AtomNet’s ability for drug discoveryAtomwise is leveraging the power of AI in an attempt to revolutionize small molecule drug discovery

Lightning AI, creator of the PyTorch Lightning framework, secured $50 million in funding from Cisco Investments, J.P. Morgan, K5 Global, and NVIDIA, totaling $103 million. With 240,000 users across 2,000 organizations, Lightning AI offers cloud-based development environments that simplify AI development. The platform integrates with popular ML tools and provides flexible pricing, including a free tier. It helps enterprises reduce infrastructure setup time and streamline AI deployment.