Enterprises across the U.S. and in key innovation hubs like Las Vegas, New York, Chicago, and Dallas are asking the same question after AWS re:Invent 2025: how do we turn AI and cloud into measurable business value, fast?
Grounded in real customer conversations and insights from the Las Vegas keynotes, This recap explores Presidio’s top takeaways from re:Invent 2025. From an AWS Marketplace that now powers large-scale transformation, to the end of “pilot-only” AI, to the moment agentic AI finally met modern cloud infrastructure.
1. Transformation > Transaction: AWS Marketplace is All Grown Up
AWS Marketplace has rapidly transformed from a transactional software catalog into a powerful engine for organizational modernization. Today, it’s where enterprises streamline cloud adoption, accelerate AI initiatives, and unlock new operating models without the friction of traditional procurement cycles. Marketplace now plays a strategic role in helping CIOs and technology leaders unify purchasing, optimize spend, and deploy validated cloud solutions at the speed their businesses demand.
Presidio has been at the forefront of that shift. By pairing deep AWS engineering expertise with Marketplace-native accelerators, we help customers move beyond simple procurement to true cloud and AI transformation. This year at AWS re:Invent, that impact was recognized as Presidio was named AWS Marketplace Partner of the Year. The award highlights how our approach is reshaping what organizations can achieve – reducing complexity, increasing agility, and turning Marketplace into a strategic lever for innovation.
As customers prepare for 2026 priorities, Presidio and AWS Marketplace together offer a path to faster value, stronger governance, and AI-ready cloud operations.
2. Pilots Fly Blind: It’s Time for Outcome-Driven AI
One theme came through loud and clear at re:Invent: the era of endless AI pilots is coming to an end. Organizations can’t afford siloed experiments, disconnected proofs of concept, or “science projects” that never make it past slideware. The mandate going into 2025 and 2026 is unmistakable: AI must deliver measurable, enterprise-level outcomes.
At its core, AI is a tool. A powerful, rapidly evolving tool, yes – but still a tool. And like any tool, it only creates value when it’s pointed at a meaningful business objective: improving patient or student experience, monetizing existing data assets, reducing operational friction, predicting failures before they disrupt operations, or accelerating time-to-insight for key decisions.
This shift from experimentation to execution is exactly why Presidio leads with a human-centered AI approach. Technology only matters when it augments people, enhances workflows, and aligns to clear organizational goals. Our industry-specific AI accelerfators – from healthcare and public sector to manufacturing, financial services, and media – help customers skip the guesswork and move straight to outcomes. They bring prebuilt patterns, domain expertise, and battle-tested architectures that turn AI ambition into value in weeks, not years.
The takeaway: stop flying blind with pilots. Start building AI that performs with purpose.
3. Agentic AI and Cloud Infrastructure Finally Meet
For years, Presidio CTO Rob Kim’s followers have turned to his re:Invent reactions on LinkedIn to go beyond the keynotes and into what new announcements, innovations, and programs actually mean for AWS customers and partners. Read some samples of his Agentic AI dispatches from re:Invent 2025 below.
AWS re:Invent 2025 Keynote Summary: Accelerating Innovation for the Developer
This year’s AWS re:Invent in Las Vegas was billed as the year that agentic AI and cloud infrastructure finally met in the middle. From CEO Matt Garman‘s shotclock timed keynote to deep dives with customer testimonials from Sony, Adobe and Writer, the keynote was packed with announcements that push the limits of what developers, enterprises and innovators can build. Below is my take on the highlights, let’s dive in.
Agentic AI and developer tools
…
AWS re:Invent Day 2 Keynote Summary: How AI Actually Transforms Work
Pacing back and forth, Swami Sivasubramanian opened his AWS re:Invent Day 2 keynote with a simple story: building a program for the first time – typing commands into a machine, waiting for output, debugging line by line. The computer didn’t remember what he tried before. Every run was a reset. Every mistake had to be debugged.
Then he stopped – now back at the center of the stage – and turned back to the audience.
“What if the systems we build today could actually remember what they’ve done?”
For me, that question WAS his keynote. Day Two wasn’t about faster chips or bigger models. It was about systems that learn through experience, persist through time, and operate continuously. In short: memory, orchestration, and durability.
Episodic Memory: When Agents Stop Being Stateless
…
