Boosting Innovation and Cutting Costs Through Lockheed Martin's AI Factory

Overview

Lockheed Martin centralized compute resources, MLOps tools and best practices into the Lockheed Martin AI Factory, built on an NVIDIA DGX SuperPOD reference architecture, to build and deploy trustworthy AI at scale on-premises under strict data governance requirements. Developers can now get GPU-backed environments running in minutes instead of weeks, and training times dropped from weeks to days. The AI factory processes over one billion tokens per week and now serves 7,000 engineers and developers, supporting internal chatbots and coding assistants such as Lockheed Martin Text Navigator, and has consolidated 30+ models on-premises.

The challenge

Lockheed Martin's AI teams faced fragmented processes, limited automation, and separate compute resources, leading to inefficiencies, increased costs and outdated compute resources, with duplicate work and inconsistent practices stemming from a lack of a unified AI strategy and clear vendor management. Setting up a development environment on local laptops or workstations under stringent security constraints often took several weeks.

The solution

Lockheed Martin established the Lockheed Martin AI Factory, centralizing compute resources, MLOps tools and best practices via a customized MLOps platform built on the NVIDIA DGX SuperPOD reference architecture for training and inference, enabling teams to build and deploy trustworthy AI at scale on-premises. Training and inference are conducted on the DGX SuperPOD, utilizing NVIDIA Triton Inference Server and NVIDIA NeMo Framework (included with NVIDIA AI Enterprise) along with CUDA. The AI Factory supports various generative AI workloads, RAG, fine-tuning of LLMs and internal chatbots and coding assistants such as Lockheed Martin Text Navigator.

Reported business value

Developers can now get GPU-backed environments running in minutes instead of the several weeks it previously took, and training times dropped from weeks to days. The AI Factory processes over one billion tokens per week. Today 7,000 engineers and developers access the AI Factory, producing thousands of automated pipelines and millions of weekly API requests. More than 1,100 trainees have enhanced their AI knowledge through NVIDIA's Deep Learning Institute (DLI), and the company has consolidated 30+ models on-premises.