Published by
Vikram Srinivasan
Published on
June 8, 2026
.jpg)
By Vikram Srinivasan, Vice President of Innovation and New Products, Optilogic
I've been in enough customer conversations over the past year to know that something has shifted. Not just in what supply chain teams are being asked to do — but in what they know is possible and feel frustrated they can't yet access.
The old workflow is familiar to everyone in this industry. A business question comes in. Weeks go into data wrangling. A model gets built. A handful of scenarios get run. Results go into a deck, get presented to leadership, and then come right back with new assumptions to test. Call it six to eight weeks, minimum, before a decision gets made — if it gets made at all.
That cycle worked when the world moved slowly enough to tolerate it. It doesn't work anymore.
The conditions hitting supply chains right now aren't arriving one at a time. A key supplier can go dark without warning. Demand can change faster than your planning sequence. And we all know that tariffs are always changing.
This is the moment we built Ada for.
Ada is Optilogic's agentic AI for supply chain design. The system is accessible via a chat interface — you describe what you need in plain language and Ada gets to work. But what's happening underneath is what makes it genuinely different.
.png)
Bring your data — messy, incomplete, real-world imperfect. Ada maps raw data to a supply chain-native modeling ontology, then cleanses it, fills in what's missing, and builds you a working baseline model — powered by DataStar underneath. From there, you add scenarios, ask questions, dig into outputs. The entire workflow that used to take weeks runs in a day. And you're steering the whole time — you can guide Ada step by step or let it run and review the work. Every step is visible.
The name Ada is a tribute to Ada Lovelace, recognized as the world's first computer programmer. What I find remarkable about Ada Lovelace isn't just what she built — it's how far ahead she was thinking. In the 1840s, she understood that computing machines could go beyond calculation, that they could compose music, solve complex problems. That vision was a century ahead of its time.
We chose that name intentionally. Because what we're trying to do with supply chain design — making it accessible, dynamic, and capable in ways people didn't think were realistic — requires exactly that kind of thinking. A belief that the art of the possible is much bigger than the current state suggests.
One thing I want to share about how Ada came to be, because I think it actually explains what makes it different.
When we started this project, we didn't ask "where can we add AI?" and we didn't ask "how do we automate what we're already doing?" We asked a harder question: if we took everything we know — a hundred-plus years of collective experience across this organization, hundreds of customer projects, thousands of hours of understanding what real supply chain problems look like — and we had to rethink the workflow completely, what would it look like?
That's a genuinely open question. And it requires a different kind of thinking to answer. Not incremental, not additive — divergent. What's the art of the possible if we start from scratch?
There was some healthy skepticism at the start, as there should be with any big question. And then sprint by sprint, it came together. The first data cleansing workflow. Then schema mapping. Then a full model build. At each step, what felt ambitious became real, and the team kept raising their own bar. Six months later, you're looking at something that can take an end-to-end supply chain design workflow — data through output analysis — and run it through an AI agent that works alongside your team.
None of this would have been possible without the customers and partners who joined us early. The people who shared their real problems, tested functionality before it was polished, identified areas that needed attention, and helped shape the use cases that made Ada what it is today — that partnership is baked into every part of Ada. Building something this new requires people willing to go on the journey with you, and we had that.
I also have to give enormous credit to the people who built this. Our AI engineers and data scientists who architected the system. Our professional services team, who've done hundreds of real projects and brought that domain knowledge into the agent design — and I can't overstate how important that is. Our product team for making something architecturally complex feel simple in the hands. Our engineering team for building it to scale. Every one of them was essential, and the energy in those sprint reviews was something I'll remember for a long time.
Speed is the obvious thing to talk about, and the speed is real. But I think it undersells what Ada actually changes.
When more people across your organization can access optimization — not just strategic modelers, but planners, analysts, even executives asking questions directly — you start answering questions you never had time to ask. That compounds quickly. Instead of three or four major design questions per year, teams can engage with ten times that. More alternatives get explored. More assumptions get challenged. Decisions get made with a fuller picture of the trade-offs.
And there's something else that happens, which I think matters a lot for the people doing this work. The modeling team stops being the bottleneck and starts being the quarterback.
Right now, modeling teams spend roughly 80% of their time on data wrangling, model building, and scenario generation. That leaves 20% for the actual thinking — the strategic questions, the stakeholder conversations, the alternatives exploration that creates real value. Ada flips that. The heavy lifting gets handled. The modeling team's expertise gets deployed where it should be: in the room with decision-makers, shaping the questions, not just answering them.
I call this the shift from a center of excellence to a community of excellence. The expertise doesn't go away — it gets amplified. More people get access to answers through composable apps built on top of the platform. And the people who know supply chain best become strategic advisors to the business instead of a service center for it. Gartner puts it well: the most impactful results from AI won't come from headcount reduction, but from human-AI collaboration — AI supporting and augmenting decision-making while people focus on higher-order strategy. That's exactly the model Ada is built on.
There's a lot of agentic AI activity across supply chain right now. Most of it is focused on execution — automating order fulfillment, routing transactions, managing operational workflows. That's genuinely valuable. But it's a different problem.
What Ada does is a step above the execution layer. We're not automating what already exists. We're helping teams reach a better state — across every horizon, from a daily routing decision to a greenfield location decision to how you navigate an acquisition. And we're doing it holistically, across the entire network using Cosmic Frog, rather than silo by silo the way most planning systems are built.
For supply chain design specifically, I'm comfortable being direct: there isn't another agentic AI capability in the market that does what Ada does.
That's not just a product claim — it's a reflection of what it actually takes to build this well. You need the optimization engine underneath. You need the domain expertise baked into the agents. You need the data platform to handle the real-world messiness of enterprise supply chain data. And you need the experience of having done hundreds of actual projects to know what questions customers are really trying to answer. That combination took years to build. Ada is what you get when all of it comes together.
I've been thinking about this a lot lately. The supply chain teams that figure this out first — that move from a quarterly modeling cadence to a dynamic, always-on design capability — will have a compounding advantage. Not just in individual decisions, but in their ability to adapt and respond as conditions continue to change. It's worth noting that a recent Gartner survey of 509 supply chain leaders found that AI and agentic AI are expected to be the most impactful drivers of supply chain performance over the next three years. The window to get ahead of this is now.
That's what this moment is about. Not faster modeling. A fundamentally different relationship between your team and the questions they can answer.
Ada is built for that.
If you want to see Ada in action, you can create a free account or learn more about Optilogic's agentic AI capabilities.
Vikram Srinivasan is Vice President of Innovation and New Products at Optilogic, where he leads product strategy across supply chain design transformation.
By Vikram Srinivasan, Vice President of Innovation and New Products, Optilogic
I've been in enough customer conversations over the past year to know that something has shifted. Not just in what supply chain teams are being asked to do — but in what they know is possible and feel frustrated they can't yet access.
The old workflow is familiar to everyone in this industry. A business question comes in. Weeks go into data wrangling. A model gets built. A handful of scenarios get run. Results go into a deck, get presented to leadership, and then come right back with new assumptions to test. Call it six to eight weeks, minimum, before a decision gets made — if it gets made at all.
That cycle worked when the world moved slowly enough to tolerate it. It doesn't work anymore.
The conditions hitting supply chains right now aren't arriving one at a time. A key supplier can go dark without warning. Demand can change faster than your planning sequence. And we all know that tariffs are always changing.
This is the moment we built Ada for.
Ada is Optilogic's agentic AI for supply chain design. The system is accessible via a chat interface — you describe what you need in plain language and Ada gets to work. But what's happening underneath is what makes it genuinely different.
Bring your data — messy, incomplete, real-world imperfect. Ada maps raw data to a supply chain-native modeling ontology, then cleanses it, fills in what's missing, and builds you a working baseline model — powered by DataStar underneath. From there, you add scenarios, ask questions, dig into outputs. The entire workflow that used to take weeks runs in a day. And you're steering the whole time — you can guide Ada step by step or let it run and review the work. Every step is visible.
The name Ada is a tribute to Ada Lovelace, recognized as the world's first computer programmer. What I find remarkable about Ada Lovelace isn't just what she built — it's how far ahead she was thinking. In the 1840s, she understood that computing machines could go beyond calculation, that they could compose music, solve complex problems. That vision was a century ahead of its time.
We chose that name intentionally. Because what we're trying to do with supply chain design — making it accessible, dynamic, and capable in ways people didn't think were realistic — requires exactly that kind of thinking. A belief that the art of the possible is much bigger than the current state suggests.
One thing I want to share about how Ada came to be, because I think it actually explains what makes it different.
When we started this project, we didn't ask "where can we add AI?" and we didn't ask "how do we automate what we're already doing?" We asked a harder question: if we took everything we know — a hundred-plus years of collective experience across this organization, hundreds of customer projects, thousands of hours of understanding what real supply chain problems look like — and we had to rethink the workflow completely, what would it look like?
That's a genuinely open question. And it requires a different kind of thinking to answer. Not incremental, not additive — divergent. What's the art of the possible if we start from scratch?
There was some healthy skepticism at the start, as there should be with any big question. And then sprint by sprint, it came together. The first data cleansing workflow. Then schema mapping. Then a full model build. At each step, what felt ambitious became real, and the team kept raising their own bar. Six months later, you're looking at something that can take an end-to-end supply chain design workflow — data through output analysis — and run it through an AI agent that works alongside your team.
None of this would have been possible without the customers and partners who joined us early. The people who shared their real problems, tested functionality before it was polished, identified areas that needed attention, and helped shape the use cases that made Ada what it is today — that partnership is baked into every part of Ada. Building something this new requires people willing to go on the journey with you, and we had that.
I also have to give enormous credit to the people who built this. Our AI engineers and data scientists who architected the system. Our professional services team, who've done hundreds of real projects and brought that domain knowledge into the agent design — and I can't overstate how important that is. Our product team for making something architecturally complex feel simple in the hands. Our engineering team for building it to scale. Every one of them was essential, and the energy in those sprint reviews was something I'll remember for a long time.
Speed is the obvious thing to talk about, and the speed is real. But I think it undersells what Ada actually changes.
When more people across your organization can access optimization — not just strategic modelers, but planners, analysts, even executives asking questions directly — you start answering questions you never had time to ask. That compounds quickly. Instead of three or four major design questions per year, teams can engage with ten times that. More alternatives get explored. More assumptions get challenged. Decisions get made with a fuller picture of the trade-offs.
And there's something else that happens, which I think matters a lot for the people doing this work. The modeling team stops being the bottleneck and starts being the quarterback.
Right now, modeling teams spend roughly 80% of their time on data wrangling, model building, and scenario generation. That leaves 20% for the actual thinking — the strategic questions, the stakeholder conversations, the alternatives exploration that creates real value. Ada flips that. The heavy lifting gets handled. The modeling team's expertise gets deployed where it should be: in the room with decision-makers, shaping the questions, not just answering them.
I call this the shift from a center of excellence to a community of excellence. The expertise doesn't go away — it gets amplified. More people get access to answers through composable apps built on top of the platform. And the people who know supply chain best become strategic advisors to the business instead of a service center for it. Gartner puts it well: the most impactful results from AI won't come from headcount reduction, but from human-AI collaboration — AI supporting and augmenting decision-making while people focus on higher-order strategy. That's exactly the model Ada is built on.
There's a lot of agentic AI activity across supply chain right now. Most of it is focused on execution — automating order fulfillment, routing transactions, managing operational workflows. That's genuinely valuable. But it's a different problem.
What Ada does is a step above the execution layer. We're not automating what already exists. We're helping teams reach a better state — across every horizon, from a daily routing decision to a greenfield location decision to how you navigate an acquisition. And we're doing it holistically, across the entire network using Cosmic Frog, rather than silo by silo the way most planning systems are built.
For supply chain design specifically, I'm comfortable being direct: there isn't another agentic AI capability in the market that does what Ada does.
That's not just a product claim — it's a reflection of what it actually takes to build this well. You need the optimization engine underneath. You need the domain expertise baked into the agents. You need the data platform to handle the real-world messiness of enterprise supply chain data. And you need the experience of having done hundreds of actual projects to know what questions customers are really trying to answer. That combination took years to build. Ada is what you get when all of it comes together.
I've been thinking about this a lot lately. The supply chain teams that figure this out first — that move from a quarterly modeling cadence to a dynamic, always-on design capability — will have a compounding advantage. Not just in individual decisions, but in their ability to adapt and respond as conditions continue to change. It's worth noting that a recent Gartner survey of 509 supply chain leaders found that AI and agentic AI are expected to be the most impactful drivers of supply chain performance over the next three years. The window to get ahead of this is now.
That's what this moment is about. Not faster modeling. A fundamentally different relationship between your team and the questions they can answer.
Ada is built for that.
If you want to see Ada in action, you can create a free account or learn more about Optilogic's agentic AI capabilities.
Vikram Srinivasan is Vice President of Innovation and New Products at Optilogic, where he leads product strategy across supply chain design transformation.
Fill out the form to unlock the full content
.jpg)
.png)
