We’re all ears when Nvidia rolls through. Jensen Huang’s confidence, matched with Nvidia’s far reaching innovation have this time introduced Rubin AI superchips, and the GeForce RTX 5090 GPU. While we don’t usually dive deep into the technicals, these innovations are Nvidia’s key to large scale agentic AI. In fact, they may become the backbone to agent run project delivery. Also, revealed at the event was Nvidia’s WALL-E called “Blue”, so nostalgic that we’ve covered it in its own story in our robotics section.
Agentic AI: Beyond a Simple Tool
AI agents are multifold relative to single LLM’s – and so the compute demand is intensive. Add in the evolution of independence in workflows (agents controlling their own activity), the demand grows. Rubin AI Superchips and the GeForce GPU are built to meet these demands not just now but in the foreseeable future. Here’s a bit about them:
- Rubin AI Superchips:
- Performance: Integrated GPU-CPU design achieving up to 3.6 exaFLOPS.
- Project Impact: Enhances real-time risk assessments, resource allocation, and strategic planning for large-scale projects.
- GeForce RTX 5090 GPU:
- Performance Boost: Doubles the performance of the RTX 4090.
- On-Site Application: Enables real-time analytics and decision-making at remote or bandwidth-limited locations, perfect for dynamic field operations.
Transforming Project Delivery
The new chip and GPU will unlock new capability, but what might this look like in our projects?
- Autonomous Task Management:
Routine tasks like scheduling and reporting can be automated end-to-end. Agents won’t need a manual trigger to start the workflow. It could be time or event triggered. - Scenario modelling:
Advanced computing allows immediate simulation of complex scenarios, supporting dynamic responses in varied environments. - Localised Intelligence:
With the RTX 5090, teams can perform detailed inspections and safety assessments directly on-site without relying on central data centres. Security and speed at your service. - Accelerated Timelines:
Rapid model training and inference from Rubin superchips enable projects to iterate in hours rather than weeks, reducing downtime and uncertainty. - Enhanced Risk Management:
Continuous monitoring by agentic AI provides proactive risk identification and adaptive forecasting. If the speed picks up, AI should be able to learn within projects not just across them. - Evolving Roles:
As AI takes over routine work, project teams shift focus towards governance, oversight, and strategic alignment—ushering in roles like AI governance officers.
Strategic Insights for Project Leaders
- Embrace Early Adoption:
Initiate pilot projects to integrate agentic AI into existing workflows and unlock new efficiencies. - Invest in Upskilling:
Equip your team with the skills to manage and supervise these advanced systems, ensuring both technical fluency and ethical oversight. - Plan Future Infrastructure:
Align your resources with the emerging capabilities of these technologies to stay at the forefront of innovation.
🥑 Food for Thought
Can agentic AI, empowered by cutting-edge hardware like the Rubin superchips and RTX 5090, shift project management from routine oversight to strategic leadership?
✅ Our Take
We believe it’s time to experiment before the complexity grows. Pilot agentic AI in your workflows to boost efficiency, enhance risk management, and empower your team to focus on high-level strategy.
The Rabbit Hole
Watch the key highlights from Nvidia’s GTC 2025 Key Note in 16 Minutes
