Breaking News: AI Agents Transform Enterprise Development
NVIDIA has deployed OpenAI’s latest frontier model, GPT-5.5, through its agentic coding application Codex, yielding dramatic productivity improvements across the company. Over 10,000 NVIDIA employees in engineering, product, legal, marketing, finance, sales, HR, operations, and developer programs are already using the system.
Debugging cycles that once took days are now completed in hours. Complex, multi-file codebase experiments that required weeks can be turned around overnight. Teams are shipping end-to-end features from natural-language prompts with stronger reliability and fewer wasted cycles than earlier models.
“Let’s jump to lightspeed. Welcome to the age of AI,” wrote Jensen Huang, NVIDIA founder and CEO, in a company-wide email urging adoption of Codex. Employees describe the results as “mind-blowing” and “life-changing.”
Groundbreaking Economics Enable Enterprise Scale
GPT-5.5 runs on NVIDIA’s GB200 NVL72 rack-scale systems, which deliver 35x lower cost per million tokens and 50x higher token output per second per megawatt compared with prior-generation infrastructure. These economics make frontier-model inference viable at enterprise scale.
“OpenAI’s stunning progress is just the latest example of NVIDIA’s work with every frontier model company,” a company spokesperson said. “Not just to accelerate the use of AI agents inside NVIDIA, but to help partners build the best, lowest cost, and most power efficient models for everyone.”
Background: A Decade of Collaboration
The GPT-5.5 launch and Codex rollout reflect more than 10 years of partnership between NVIDIA and OpenAI. The collaboration began in 2016 when Huang hand-delivered the first DGX-1 supercomputer to OpenAI. Today, Codex harnesses GPT-5.5 through code analysis, generation, and debugging capabilities.
To ensure security, each agent operates in its own dedicated cloud virtual machine (VM) via remote Secure Shell (SSH) connections. NVIDIA IT rolled out VMs for every employee, providing a sandbox with full auditability. A zero-data retention policy governs the deployment, and agents access production systems with read-only permissions through command-line interfaces and Skills—the same agentic toolkit used for automation workflows across the company.
What This Means
This deployment marks a paradigm shift in how AI agents assist knowledge work. By processing information, solving complex problems, and generating new ideas, Codex with GPT-5.5 moves beyond simple code completion to true agentic innovation. The combination of enterprise-grade security and 30x cost-performance improvements makes this model practical for large-scale corporate use.
For NVIDIA, the immediate benefit is accelerated product development cycles. For the broader industry, it signals that frontier models can now be deployed securely and economically inside enterprise firewalls. Other companies are likely to follow suit, leveraging similar infrastructure to empower their own knowledge workers.
“Every agent needs its own dedicated computer,” the spokesperson noted. “We’ve created a secure environment where agents can operate at maximum capabilities while maintaining full auditability for IT and compliance teams.”