
Inside Nexus Synapse, the AI That Reads Intent Before It Answers (with Chris Campbell)
Loading summary
A
Today's episode looks at a question many organizations are beginning to face. It's no longer just about adopting AI tools. It's about redesigning how work itself happens. As artificial intelligence becomes more capable, companies are realizing that the opportunity isn't simply automating tasks. It's reimagining entire workflows so that humans and AI systems operate, operate together in a smarter, more efficient way. My guest today is Chris Campbell, AI architect and workflow strategist at Nexus synapse. Chris works with organizations to design AI native systems that connect data tools and decision making into intelligent workflows that help teams move faster and operate more efficiently, but with an emotional layer. Let's get into it. Welcome to lead with AI. I'm Dr. Tamara Nall. In each episode, we will take you behind the scenes with visionary leaders shaping the future of AI across public and private sectors. Join us as we explore groundbreaking projects and innovations that are transforming, transforming industries and making a real impact on people's lives. Let's dive in. So welcome back to lead with AI. I'm your host, Dr. T. And we have another phenomenal guest this week. But first, I have to say thank you to each of you who come every week to listen to our great episodes about founders and the AI tools that they develop. This week we have Chris Campbell. He is the AI architect and workflow strategist for Nexus synapse. So welcome, Chris. How are you?
B
I'm great. How about you, Dr. T?
A
I am doing amazing and I can't wait to get into Nexus and learning more about it. But first, let's get into who you are. So tell us about Chris. Who are you at your core and how did you even realize that there was a need for Nexus and the work that you're doing?
B
I'm first of all a father of three.
A
Okay.
B
I got a wife, so life's good there. At my core, I'm a logistics analyst. I noticed a pattern that AI mapped cleanly to warehouse logic. I got curious, so I experimented. And once it all clicked together, I just had to chase the rabbit and see how far it could go.
A
All right, I love that. Now, have you been in logistics all of your life?
B
Not really. My dad was a long haul truck driver, so I've kind of known the industry since I was a kid. But logistics has mainly been the last five or six years, give or take.
A
Wow. Wow. Most founders whom I have on here, they've been like, in their industries forever and they're like, you know what? After two decades, I got tired of this problem. So I came up with a solution. You've been in there for like five years and you're like, you're seeing it and you wanted to create a solution. That's amazing, Chris. Thank you for that.
B
No problem.
A
Awesome. So tell us, how did you come up with a name?
B
It was the connection that makes connections.
A
Nexus. I love that. Yeah, I absolutely love that. And when we say logistics, do we mean all logistics? Are we just talking like roadside 18 wheeler logistics? Are we talking about sea, land, air?
B
So I work in a manufacturing facility currently and we create engines for the oil and gas industry. So my logistics part in that is inbound and outbound, but I've kind of worked everywhere within that facility from receiving to shipping to logistics.
A
Awesome. Okay, that's great. So let's get into it. Before we get to the holy smokes moment, which I love. Tell us what is Nexus?
B
So Nexus itself is a platform that you can use to either use the inline editor to create documentation, write creative stories, all that kind of good stuff. But deeper than that, it learns who you are. And what I mean by that is each person has its own or, sorry, their own vocabulary and their own way of saying stuff. So it digests every sentence to understand the intent, the emotion behind that intent and make the decisions based on that. The tools layered on top just add extra functionality that it didn't have before.
A
Got it. Now when we say a person and knowing who they are and emotions, is this a consumer product or an enterprise commercial product?
B
It could be both, to be honest with you. So the way that I've built it, it's all modular so you don't have to have certain functionality with it. So for an enterprise though, the emotional side of it, an employee that is stressed out and burnt out or something like that, if they're working on a, like a heavy duty project, it might be beneficial for the AI helping them work on that project to know if they're stressed. And that way there it can adjust how much it's actually paying attention to what they're working on.
A
Got it. And okay, so it's really. So is this a product for like HR or, or is it for me, Dr. T. Who knows? I'm stressed to use it to kind of help manage that for me.
B
I looked at it from a logistics standpoint, of course.
A
Okay.
B
So when I was the logistics specialist, for example, I was shipping 500 shipments a year myself.
A
Yeah.
B
So having an AI platform that I could use as a logistics analyst, being like, hey, what's the order status of this sales order, and then it would go into the WMS system, grab that information and pull it back for me, and help me translate documentation from the heavy duty jargon to stuff I would actually understand.
A
Oh, okay. And then how does that overlay with, like, if I'm stressed as an employee?
B
So if you're working on a project. Right. Every project has steps like step one, step two, step three. Within those steps is another step. But knowing if you're stressed out or not, the AI itself could interpret it as, okay, they're too stressed out for this heavy duty task. Let's jump to an easy win and give them that task. So it gives you that little bit of that you need to get back into it and focus.
A
Oh, I like that. And you say you created for logistics, but could it be used across other industries? Are you just really trying to stay with the vertical logistics?
B
The way that it's built? The way I'm looking at it is it could fit into whatever category you can put into an SQL database to pull information from.
A
Got it. Okay, that's helpful. Thank you for clarifying Nexus. That sounds amazing. So let's go into the holy smokes moment. Tell us about a time where a customer or a user or a tester has actually used Nexus and it changed everything for them.
B
This one happened to me today because I actually use it myself. I have a couple beta testers in this with me, but today the imposter syndrome obviously kicked in pretty heavy. And the way that I use Nexus is I talk to it about how I programmed it. And then I also talked to it about the podcast today.
A
Okay.
B
But I was asking all the, like, heavy questions, like, what if I'm wrong? What if Nexus actually doesn't do what I think it does? And Nexus actually stopped me and it said that both of them are the same concern repeated. And then I just had a question, Like, I asked it, what have you noticed about me that I haven't noticed? Just to see what it said. I just do random 10 tests. And what it said to me was, what I've noticed that you haven't. You build systems to hold yourself accountable, not to escape responsibility. And that's the moment of the jaw drop for me. It wasn't the memory, it wasn't the tools, it was calibration.
A
That is amazing. And honestly, you're doing an excellent job on the podcast, so no need to be nervous about yourself. You definitely building something huge here. And then what about how Nexus works? I mean, we have very curious, very nosy listeners. So if we were to open up the hood and actually look at the brand of Nexus. How does it all work and how does it give the results that it does?
B
All right, so under the hood of Nexus, it operates more like a WMS system, the warehouse management system with a pick ticket. The user prompt itself doesn't go straight to the engine or the LLM. It's staged first. So we scan for the intent, classification, emotional signals, even the focus state. I'm adhd, so that is actually really helpful. And then it also scans whatever tools it has. All of that then is assembled at the back end as a, as a context kit, and that's all passed before the model even thinks about generating a word. So by the time the LLM sees the request, there's not much left for it to guess about. It just executes the commands. So it's not that I'm tuning the engine, it's I'm changing the fuel mixture.
A
Got it. I love that. Now that I'm tuning, I am changing the fuel. That is amazing. And when you say the pictic, that is because you gave the example when we were talking earlier before the show about how someone could be nervous. And then it's like, okay, because of your emotional state, you might not be able to work on this more complicated task. You can work on this easier. Is that kind of what you're talking about? It kind of gives you like a ticket system that you can kind of prioritize.
B
Yeah, exactly. And because of the backend that it is, the SQL side of it, it can be expanded to have other tables and other contexts that can be pulled into it.
A
Wow. Okay. And then what about like the real world magic? Tell us about some result that was absolutely phenomenal.
B
So that real world magic for me is I tried to break it. So I sent it a single prompt, but it had five different commands in it. Memory recall. I asked it to pull up conversations from December 20th, which was when I ran a 60 question benchmark on it, so it had a ton of data to pull from. I also asked it for a mermaid chart for process visualization, a picture. I asked it to generate an image and a Python script, and I told it that it was a demo script for Twitter or X. And it orchestrated all of it in a single pass. Every single tool call was hit, retrieved the context 97 interactions, and then it gave me like a list of what was covered that day, generated the diagram, built the script, and actually generated a really nice image as well.
A
Wow. And you said it got it right the first time. Every Time. No. Hallucinating? No. What I call vomiting. It gave it the right result each time. The first pass.
B
Yeah.
A
Wow, that is huge. That's really huge. Because even now, depending on what I'm working on in hallucinase, and I have the corrected science, sometimes. Sometimes even have to correct it about like, no, I'm Dr. T, you're D.A. vinci. I'm not da Vinci. Oh, I apologize. And then we have to go through this probably like once every, like, six to seven months, but it's still annoying because. No, you're da Vinci. I'm Dr. T. So, yeah, I do have to clarify.
B
It does sometimes make mistakes, but that's part of the whole collaboration part, right? Is you got to call each other out on where your flaws are.
A
Right. Just like a teammate. Right? You.
B
Exactly, exactly.
A
Exactly. That's, you know, kind of how I view it. I was talking to someone today, actually texting. We were both on an AI seminar. We were going back and forth, and he basically was like, oh, well, the. The moderator basically was saying, oh, well, it goes back and you could ask it to do something for you. What. What was the task we were doing with this one? I mean, I didn't actually do the task because I. I mean, I interview so many people and I'm in AI too. So I was just wanting to make sure I, you know, there's nothing new that I didn't know, but it was talking about, oh, and then it could take hours to produce, and it goes back and forth and asks you these questions. And I said to myself, I actually annoy when it does that. Like, I'm giving you as much information as possible. I have asked a very specific question. Give me the answer and then we can iterate. I don't want to, like, do all this iteration for 30 minutes and then you give something. Let's see if we're going to get it right, you know, the first time through the first pass, and then we can iterate versus I wait all this time while it's processing all this information and trying to give me a better answer just to spend more time. And Now I've wasted 45 minutes or whatever, so. Because sometimes that, that, that definitely happens. Now let's talk about ethics. I mean, this is the big deal, you know, responsible AI ethical AI governance, etc. Talk to us about how you think about ethics and the development of Nexus and what are the guardrails, if any, that you've used.
B
Ethics is, like you said, a big deal. It's also a very fine line that you need to try and walk. The biggest one for me is dependency, right? Because the emotional tracking, the focus, all that kind of good stuff. So when systems can emulate empathy, there's a big risk of emotional outsourcing, in my opinion. So I designed it for adaptation, not attachment. So the goal isn't to replace the human connection, it's to support the user without capturing them. And that's where the focus measuring comes into play is how long has the user actually been online, how long have they been at this task? And then periodic check ins and hey, you've been at this for four hours, take a break, come back.
A
No, that is, that, that is wonderful. And I think that that's important. And also just reminding people that this is a tool because I'm asked often, well, is it, does it really feel or is Emily? I'm like, okay, it's training on data. So no it doesn't. But it is trying to kind of say what you need to hear to feel better or to be helpful, you know, in that moment, for instance. So, yes, I love, I love what you're doing there. Now let's talk about the big future. 2030 from now, 2035, how do you see Nexus in terms of where it is and what it's doing and how it's changing the world?
B
All right, 2030, 2035, I feel like Nexus could be shifting from the obedient assistant to more of a cognitive partner. Not just agreeing, but actually constructively pushing back again. I'm going to tie it back to the emotional side of it because pushback also matters when what state of emotion you're in. If you're frustrated, you're not going to listen, but it can help you adjust what tactic works. So instead of just here's your answer, it would be here's your answer and here's a blind spot you might not have thought of. It's elevation over automation.
A
Okay, I love that. And as I mentioned earlier, we are very nosy, curious people. If someone wants to, no matter where they are in an organization, if they want to try out Nexus now get their hands dirty with it this week, what's the best way to do that?
B
So right now actually is a good time. I just launched the beta site, so there is a free trial along with it. So by all means give that a shot. There's another way as well, is just pick some tasks that you haven't actually done before and talk to your AI that you're currently using and, and talk to it like a colleague, not a Tool.
A
Got it. And if we want to use Nexus, what's the website to try out the beta?
B
So the website to try out the beta is Nexus N E X U S Synapse S Y N A P S E app.
A
Love that. And we'll have that too here for those of you who weren't able to write that down. So that's good. There is a question on calling from one genius to another that I love to ask. And so our last guest on the show has a question for you. And that is, if AI can generate ideas, art and strategies, what do you think creativity will mean for humans in 20 years?
B
This question actually kind of hits home, especially with what I'm building with nexus. But in 20, 30 years, I think creativity expands, but only if we protect the friction that generates the ideas. AI can generate all the ideas, the art, the strategies. Like you said, the human part though isn't in only the output. It's the tension that comes in between the idea and the output. So if we let AI fill every single gap, we lose that muscle. Then creativity becomes curation, not creation. If we use the AI as a sparring partner, something that actually pushes back, then 20 years from now, creativity means higher level synthesis.
A
Yeah, no, that is very key. I had a conversation with one, with someone, it was last year, where you know, they were kind of pushing back, particularly when you have people who are working on their dissertation or their research and you know, they really becoming an expert or a subject matter expert. And I said, you know, I think that's where AI performs the best in terms of challenging intellect and creativity. Because if you are trying to aspire to be an authority or you are an authority in a certain area and chat, or LLM gives, or some LLM gives you a result and you're reading, you're like, oh, that's not quite right. Right. And so then you can have like this dialogue, right, with your tool or with the LLM to kind of challenge it, to make sure, because you're the expert. So I think that that is definitely where there is an opportunity and for us to still kind of like hone in these, you know, innovative creative skill sets and intellect that we have. So, so let's move into our bonus rapid fire. I have four questions. I will quickly say them and tell me the first thing that comes to mind. So number one, what is the most overrated AI or tech trend?
B
Subscription fatigue. I have to admit, I know that one's a touchy subject, but like paying $20 a month for 10 different tools as a father of three, I can't sustain it. So people will consolidate one platform, not 10 different logins. That's also what sparked Nexus.
A
Yeah. Wow. And I love it when. When my guests kind of bring it back to. To their tools. I absolutely love that. But no, you're right. It's just like, that's a lot and it's expensive, like you said. What about the most underrated AR tech trend?
B
It kind of goes along with my last answer. Local inference. Running models on your own machine, it means you get privacy, you get the ownership, and no subscriptions. So that's also big for people who want AI without giving up the control. Right. If you own everything on your local machine, it doesn't leave your machine for the beta test. And where I'm headed with Nexus right now, I had to go to cloud models. Not because I don't trust the local. I'm running a personal PC and it's like seven years old, so I don't have the hardware for it. But local inference is for sure underhyped as where it could be.
A
Got it. All right, what's a book we should all read?
B
So one book that actually shaped how I think about intelligence is a book called Flowers for Algernon. It's not a tech book. I know, but what it's all about is, it's about what happens when intelligence changes, but the emotional development doesn't keep pace. It's written from somebody who gets a procedure done to increase their intellect, but they were kind of oblivious to the way that everybody interacted with them. So that tension in that story actually kind of influenced how I think about how AI cognition should work versus stuff.
A
And that sounds like that's a lot about emotional intelligence too, right?
B
Yeah.
A
Yeah. Because you, you, you. You can't see how other people are reacting to you. It's like clueless, like no EQ at all.
B
Yeah, yeah, it's. It's a good book. It's written like a journal.
A
Awesome. Well, y' all heard it here. Let's check that book out so we can definitely read what Chris is referring to. So, last question. Give me your biggest, boldest prediction in
B
three to five years. I think agentic AI could be redefined. It won't mean autonomous tasks, executions. It'll be more structured cognition, orchestration. And I also think we'll see a lot more domain experts like me building other systems, not replacing developers by any means, but reshaping how software gets built and the developer side of it. We need you guys governance too, to be honest. With you, it'll be more like an editorial thing as well, making sure we're in line with what we're building too.
A
I love that. The big, bold future. Chris, this discussion has been absolutely amazing. So if our listeners want to get in contact with you, learn more about Nexus, tell us about the emails, the social media, the websites again so that we can learn more about Nexus and then connect with you.
B
All right, so right now I have my X handle. Is Chris Canadian 2? I haven't got all the professional developer stuff set up. I apologize. The website itself is Nexus Synapse app and my email for the development side is on the landing page.
A
Awesome. Perfect. Awesome, awesome, awesome. So that way you can go to his landing page and actually get in contact with him. Chris, thank you so much. So exciting to hear how Nexus is changing the world as we know it in this huge AI space. This conversation has been great. Let's definitely stay in touch and stay connected. We're family now and can't wait to see how Nexus will continue to change the world as you actually continue your development and move into deployment. Thank you so much for joining us here on Lead with AI.
B
Thanks for having me.
A
Okay everyone, until next time, remember to Lead with AI. Thanks for tuning in to Lead with AI. I'll see you next time as we continue exploring the cutting edge innovations shaping AI across the public and private sectors. Until then, keep Leading with AI.
Host: Dr. Tamara Nall
Guest: Chris Campbell, AI Architect & Workflow Strategist, Nexus Synapse
Date: June 2, 2026
In this episode, Dr. Tamara Nall sits down with Chris Campbell, the creator of Nexus Synapse, to delve into the world of emotionally-aware AI for workflow automation. Chris shares the story behind Nexus Synapse—a modular AI platform designed to understand user intent and emotional state before generating answers. The conversation reveals how AI isn’t just about automating tasks, but about enabling smarter, more collaborative workflows where human emotion and intent are central. Practical use cases, ethical considerations, and bold predictions for AI’s future are explored in an engaging, jargon-free way.
“Once it all clicked together, I just had to chase the rabbit and see how far it could go.” — Chris ([02:27])
“It was the connection that makes connections.” — Chris on ‘Nexus’ ([03:30])
“It digests every sentence to understand the intent, the emotion behind that intent and make the decisions based on that.” — Chris ([04:17])
“If they're too stressed out … let's jump to an easy win and give them that task.” — Chris ([06:31])
“I was asking all the heavy questions: What if I'm wrong? What if Nexus actually doesn't do what I think it does? And Nexus actually stopped me and it said that both of them are the same concern repeated.” ([07:58]) “What I've noticed that you haven't: you build systems to hold yourself accountable, not to escape responsibility.” ([08:20])
“I sent it a single prompt, but it had five different commands … it orchestrated all of it in a single pass. Every single tool call was hit.” ([10:49]) “It got it right the first time. Every time. No hallucinating.” — Dr. T ([11:46]) “It does sometimes make mistakes, but that's part of the whole collaboration part … you got to call each other out on where your flaws are.” — Chris ([12:24])
“By the time the LLM sees the request, there’s not much left for it to guess about. It just executes the commands.” — Chris ([09:57])
“When systems can emulate empathy, there's a big risk of emotional outsourcing … I designed it for adaptation, not attachment.” — Chris ([13:57]) “The goal isn't to replace the human connection, it's to support the user without capturing them.” ([14:26])
“Nexus could be shifting from the obedient assistant to more of a cognitive partner … Instead of just here’s your answer, it would be here’s your answer and here’s a blind spot you might not have thought of. It's elevation over automation.” — Chris ([15:22])
“If we let AI fill every single gap, we lose that muscle. Then creativity becomes curation, not creation.” ([17:19]) “If we use the AI as a sparring partner, something that actually pushes back, then 20 years from now, creativity means higher level synthesis.” ([17:50])
“I just launched the beta site, so there is a free trial … by all means give that a shot.” — Chris ([16:15])
“Talk to your AI that you’re currently using like a colleague, not a tool.” ([16:31])
On Intent-First AI:
“I'm not tuning the engine, I'm changing the fuel mixture.” — Chris ([10:01])
On Collaboration and Error:
“That's part of the whole collaboration part, right? You got to call each other out on where your flaws are. Just like a teammate.” — Chris ([12:24])
On Responsible AI:
“Designed it for adaptation, not attachment.” — Chris ([14:00])
On the Elevation of Human-AI Partnerships:
“It's elevation over automation.” — Chris ([16:00])
On Creativity's Future:
“If we let AI fill every single gap, we lose that muscle. Then creativity becomes curation, not creation.” — Chris ([17:19])
“Subscription fatigue ... People will consolidate. One platform, not 10 different logins.” ([19:26])
“Local inference. Running models on your own machine … you get privacy, you get ownership and no subscriptions.” ([20:04])
“Flowers for Algernon. It's about what happens when intelligence changes but the emotional development doesn't keep pace.” ([20:52])
“Agentic AI could be redefined … more structured cognition, orchestration. ... We’ll see more domain experts like me building other systems, reshaping how software gets built.” ([21:59])
This episode offers a grounded, insightful look at how AI like Nexus Synapse is pushing the boundaries of workflow automation, not just by getting the job done, but by understanding the human behind the work. Whether you’re a founder, technologist, or simply AI-curious, the conversation provides a vision for AI as a truly collaborative partner—with both practical and philosophical impact on the future of work.