All right, so today on e speaks we're diving into a concept that sounds a little futuristic but is quickly becoming an operational reality. That's the autonomous enterprise. Modern organizations don't run on single systems anymore. They run on interconnected workflows, applications talking to data pipelines, cloud services interacting with legacy platforms, and decisions happening in real time across hybrid and multi-cloud environments. When those workflows stall or collide, it's not just a technical hiccup.
It's missed SLAs, compliance risk, lost revenue, and shaken confidence. So, the question becomes, how do enterprises move beyond simple automation and towards intelligent orchestration? To explore that, I'm joined today by Jennifer Margolis, director of product management at BMC Software. We're going to talk about how AI-driven workload automation through platforms like Control-M is shifting automation from a back-end efficiency play to a strategic business enabler.
Jennifer, welcome to e speaks. It's great to have you. Thank you so much for having me today. I'm excited to be here. Awesome. Well, I guess to get started, we've all heard the term automation for years, but BMC talks about orchestration in addition to automation. What's the real difference between traditional job scheduling and intelligent orchestration? Sure. So, your traditional job scheduling is really focusing on making sure that a task done, that it gets completed on a timetable or pretty simple schedule with simple dependencies.
Orchestration goes beyond just making sure that a job gets done at a specific time to organizing the context around jobs, the more complex environments that they work in, and the interdependencies and making sure that the right things get done in the right order at the right time. And it also allows you to bring in events to trigger your jobs or tasks or workflows instead of just having to come up with a static timetable for that, right? So, it makes it easier to manage your entire business workflows across, you know, on prem, distributed, or cloud environments and make everything work together and coordinate all your different data pipelines with your application workflows with your business, you know, services that you're trying to deliver.
Make sure that everything gets to the customer on time and in perfect condition. Okay, that makes a lot of sense. And and and leads well into this. So, enterprises today operate in hybrid and and multi-cloud environments plus legacy systems that aren't going anywhere. What has changed in the last, say, 5 years that makes AI-driven orchestration necessary and not just simply helpful? Sure. So, we've seen in the last few years that organizations are dealing with a lot more data volume now and more complex environments that their data lives, that their applications run, and the infrastructure that they need to manage.
So, all of these combined make it even more challenging to make sure that your mission-critical workflows are running on time every time and it makes it harder to track down and fix if there's an issue that goes wrong. So, it makes manual coordination more difficult. And it also means that organizations are facing higher risks with those kind of static schedules like I mentioned with the traditional schedulers. When you have a simple environment, it's a little easier to just make sure everything runs on time.
But when you have a larger environment or you have it spread across different applications, different systems, different types of infrastructure, then one small task going wrong or not running on time can really snowball into wider impacts. And it can start to impact your SLAs, your costs, your overall workflow efficiencies. And that can be really hard to spot if you're if you don't have a control plane on top or a single pane of view. So, yeah, I would say that's kind of been the big shift we've seen in the last few years is now AI-driven orchestration is bringing that intelligence and that monitoring to help predict bottlenecks, so you can prevent them, to design your workflows more efficiently both in terms of performance, but also costs.
Cuz we know, as things scale, costs increase and we definitely need to keep those under control. It's really become more essential now for managing workload automation or workflows across these hybrid environments. For sure. Well, when you talk about AI and workload automation, what what does that actually mean in practice? Great question. So, AI is a fun term and very trendy and we hear it a lot, but what does that mean in reality? Well, for AI-driven workload automation, that means using machine learning and generative AI or agents to make better decisions about how your workloads are designed, how they are scheduled, and how they're operated.
So, in modern orchestration platforms, like Control-M for example, you can use predictive analytics to forecast your SLA risks. We've got that are managing like 5,000 SLAs a day. So, looking at that manually and trying to identify where you might have problems in the future can be a little overwhelming. So, AI is really great for processing a lot of information at a higher scale like that. So, it can help flag, "Hey, this is your SLA that's at risk.
Nothing bad has happened yet, but if you want to prevent something bad from happening, you know, now's your chance. So, that's super helpful. It can also help uh detect anomalies and recommend remediation. So, if you've got a job that's starting to take longer to run for some reason, you know, it might not have tripped over yet that threshold into being like too long, but AI can help identify if there's a trend in your increasing run times that could signify an underlying issue.
And you can take action before your customers or your end business users feel the impact from this, which is really nice. We've also got AI assistants and kind of in-tool AI support and agents to help you just have a nicer user experience. So, AI can help you with troubleshooting. It can help you just create and design workflows from natural language. So, you know, back in the day when we first had dashboards being able to be created from just telling, you know, typing in, "I would like a small dashboard." Well, now you can do the same sort of thing with AI agents creating the actual workflows, which is a lot more advanced and really cool, I think.
It's really fun. It also >> It is. It opens up orchestration to more users within the business. Now, it's not relying on just more specialized, like really experienced folks. You can bring in more data ops teams or devops teams and or business teams and have them be part of that ownership over their workflows, which is really cool, I think, too. That is cool. That is cool. Let's take a look at a regulated industry, like say banking. So, imagine an overnight batch process kind of risk calculations or fraud detection fails or runs late.
How would a smart orchestration change the outcome compared to traditional scheduling? Sure. So, that's a very common instance where you've got overnight batch processing and if something fails with more of the traditional scheduling approach, you're not going to notice that until the next morning. So, you know, imagine you come into work and you've made your coffee and you're like really excited to be super productive for the day and get so many things done and then you realize, "Oh my gosh, all these like alerts and flags from overnight." Now you're just shifting into fire fighting mode and you have to deal with all that and there goes your plans for the day and now I guess tomorrow I'm working twice as fast to make up for what I didn't get done in my real job during today.
That is a more common situation with the more traditional scheduling. The nice thing about AI-driven or smart automation is that it doesn't leave problems waiting for someone to find them. You can define guardrails and policies and standards around either how to react to a problem when it surfaces. So, you can design alternative workflow paths if one breaks or you can define how many times you want to retry a job. You know, maybe it just got stuck and just needs a retry, that's fine.
Or you can identify important stakeholders. So, you know, something's mission critical and you've got 24/7 support or follow-the-sun support and you've got someone there that can fix it, maybe that's when you bring them into that uh process, right, at that point. Yeah. Smart automation also helps document everything for you. It gives you logs and documentation for every action that's taken so by the AI or by the person and helps control effectiveness and prepare you for compliance or reporting in those regulatory industries if we're taking that as an example, which is pretty nice.
And the cool thing that I'm seeing now is permission controls for AI agents within a tool. So, Mhm. I mentioned Control-M Scott, you know, AI assistant and agent that can can help you do things in the tool. You can assign them essentially like role-based access control and groups as if they were people to help put those guardrails on AI as well so it doesn't run away on you. Yeah. You You have a night time of problems compiling on top of each other.
Instead, it's able to deal with it and tell you what it did. Yeah, so you can sleep at night and like, you know, everybody needs their beauty sleep, so. That's right. Nothing should ruin our coffee. Yeah. In industries like insurance or manufacturing, workflow reliability is more than uptime. It's It's really about compliance and auditability. How does orchestration help build resilience while also supporting governance and regulatory requirements?
Absolutely. So, a lot of our customers are in regulated industries or in industries like manufacturing, where you really want that auditability, that governance, and that control. And you're going to see that enterprise organization platforms really help centralize the definition, execution, and monitoring of workflows, including the governance around them. So, that means every step in the workflow, each dependency, and all the actions that are being taken are being logged in one place, which makes it a lot easier to support your audit activities to prove that you are governing and enforcing the policies that your organization has put forth.
And all of that really detailed history, as well as the additional policy control or site standards enforcement, makes it just really easy to demonstrate that you are following these guidelines. The site standard enforcement is one particular one I really like about organization platforms like Control-M because I wish that I could have that everywhere in my life. Like, you can, for example, enforce a specific naming convention for your workflows or your job, so it's not just, you know, file transfer version two final final final.
It's like, "Who What in the heck is that mean?" Cuz that's how I roll. And it's been be nice to get past that. Yeah. So, you can set those best practices and enforce them so you're not just relying on individual people to remember and you know, police themselves essentially to oh, I got to make sure that I'm putting in a this in here or like which system this goes into and and all that. It makes it a lot easier and it removes some of that mental load from them.
It's just taken care of by the system, which is really nice. So, that and then yeah, the kind of automatic recovery or failover steps I mentioned a little bit earlier with the event-driven capabilities in that too. Like if something bad happens, then that event can trigger a recovery path to try and get everything back and help you meet those SLAs cuz it's a very important. And >> Absolutely. All of that combined just really helps eliminate those blind spots or gaps that you otherwise have to contend with if you're trying to do it all manually, right?
Yeah. Yeah. Yeah. Well, let's let's talk about data pipelines here for a second because they're becoming the backbone of AI initiatives themselves. How does intelligent orchestration ensure that data workflows execute reliably and fast enough to support real-time or at least near real-time decision-making? Right. Well, the nice thing about orchestration is that it is coordinating data workflows across different types of environments like we mentioned earlier.
You know, you've got your on-prem, your cloud, your distributed. It's also coordinating across different types of streams like batch processing or your streaming systems. So, you could have, you know, Kafka data coming in and that ability to connect across the breadth of the data, where it's living or where it's working as well as the kind of complexity of it, means that you can really make sure that you're the right data ends up in the right place at the right time to feed those analytics dashboards that you have or those AI models that then power your customer-facing services or aid in your higher-level decision-making, right?
So, we can plug in Control-M to react to messages from Kafka or AWS SQS or other services and then trigger those pipelines immediately. And then you're not waiting for a fixed schedule or for someone to come in and manually trigger it or to move the files and wait like it it can all just run a lot quicker and a lot more seamlessly and not depend on human interaction. Yeah. That's such a a massive improvement over the way that's happened in the past. >> Right.
Yes. And keeping that speed at any you know, keeping the latency low really helps reduce that issue of like data freshness. So, you're not waiting too long to process the data such that now it's like out of date. You can really make much more timely decisions. That's pretty critical in you know, some of those industries we've talked about today like banking, financial services. If you're doing like fraud checks or you're trying to serve up personalizations to your customers, it really needs to be on near real-time data.
It doesn't really help if you process a transaction and like 2 days later you're like, "Oh, sorry, that was flagged as fraud." It's like, "Well, now you've already debited it." So, like that doesn't really help. Oh, that's excellent. Makes It cuts a lot of risk out of the equation. >> Right. Right. When organizations implement AI-driven workload automation, what measurable improvements can they expect to see? Reduced SLA breaches, faster execution times, lower operational overhead?
Where does Where does that ROI show up the most clearly? Well, you've really hit on it. I mean, organizations are typically seeing fewer of those SLA violations or breaches, which is really nice because as we know, what gets measured gets improved. So, if you are able to show that you're improving on SLAs, you are advancing towards your goals, which is awesome. Everybody loves that. You're also getting, you know, typically faster end-to-end workflow completion.
So, it is by removing that manual effort and those kind of delays waiting on human action, you can just get a lot more done faster and that helps lend a bit more of a competitive edge over some of your, you know, competitors that might be still doing things a little more slowly in a more traditional scheduling approach. Uh Yeah. We're seeing customers reporting higher success rates for their critical patches. So, speed is really nice, but reliability is more important, maybe equally important, but in my opinion, more important than just speed at scale.
So, you're doing things faster, but more importantly, it's reliable and and things just aren't breaking. And that's just by removing some of that potential for introducing human error or variations uh if you're doing things more manually. So, that's the benefit of automation anywhere you're going to deploy it. Yeah. That's wild. It's such a an interesting time to to watch how this is all evolving. And and because of that, there's sometimes this fear that more automation means less human involvement.
Uh in reality, what happens to IT and operations teams as orchestration becomes more intelligent? Right. Well, we're not removing the humans altogether. It's a very valid fear and you hear people talking about, you know, it's "They're going to take my jobs" and things like that. Totally valid. What we're noticing in reality though is that the work that people are doing is shifting more from that reactive, firefighting mode, that low-level, kind of repetitive task work, shifting that over to the AI or the automation so that it can take care of that and instead now these people on IT teams are freed up to do more architecting, more strategic planning, spend more time understanding the goals and requirements of their business stakeholders so they can support them better instead of like chasing around failed jobs and doing rework and just trying to get things going.
So, when automation takes care of things running smoothly, the humans can do what they're best at and that's more of that creative thinking, that design, that strategy approach and that's when we get the opportunity to bring in that new ideas for more innovation, right? If you're just stuck fixing things all the time, you don't get that like joy of discovery and experimentation of like bringing in something new. Yeah, and that's that's working on the valuable stuff uh which is so important.
And and who doesn't want more time for that as opposed to the miserable part of our jobs? Right. And the approach on my team at least is we're taking a like augmented approach, right? So, we're bringing in automation, we're bringing AI not to replace people but to augment their work so that it makes their lives easier and then they enjoy their jobs more and they want to stick around and and you know, build the future with us. So, keeping them in control, keeping the people in control of the policies and the guardrails and the governance and then the approvals for high-stakes decisions is I think like a nice balance between handing off some of the work to AI and automation versus keeping it all on the plates of your team members.
For sure. For sure. I I totally agree, too. I think that's a that's it's a really big unlock that doesn't get talked about often enough. I've got one last question for you here and looking forward, what is the next stage of automation in the enterprise look like? Are we moving toward fully self-optimizing workflows or do you think there's always going to be a human-in-the-loop model? I think that there is always going to be a human-in-the-loop model, or at least always the option to choose that, right?
So, what we don't want to do is just force everyone into, you know, AI running everything and limiting the visibility into that. We still want the transparency and the visibility for humans because ultimately it's up to humans to decide where we're going, and also to be accountable for where we're going. So, I would say the next stage of a automation moving towards, like you said, the self-optimizing workflows with still a degree of control and options so that Yeah. people are defining the policies, the guardrails.
Um, they always have a flag that can stop things if things are going wrong or have that part of the automated process. I think you can see this already kind of in some of the orchestration platforms, right? >> Yeah. Where you can define access control and policies around what AI can do in your automation tool, and you still have those abilities to have like manual go no-goes on decisions, uh, even if a lot of the busy work or more tedious work was completed in the front end by the AI or the automation.
So, over time I really think we'll see much more of a blend between these AI agents that we're working with and this intelligent automation with people just really working cohesively together. Like, it will be second nature just like, I don't know, using Word to type out your documents, or even just like >> Yeah. talking into your phone to dictate to, you know, voice to text. It seems like a very natural, normal way to interact instead of having to handwrite everything out.
Uh, so I feel like we'll get used to using AI agents to support us in our work in a similar fashion pretty soon. Absolutely. I I think so, too. I think I think we're there. Yeah. You know, what stands out to me from this conversation is that automation alone isn't enough anymore. Enterprises are struggling because they lack tools. They're struggling because their workflows are fragmented across multiple environments, clouds, and legacy systems. Intelligent orchestration brings those moving parts into sync.
It predicts bottlenecks before they happen. Optimizes resources dynamically. And most importantly, it turns workflow reliability into a strategic advantage instead of a back office concern. Jennifer, thank you so much for joining us today and helping unpack what smart automation really means and looks like in practice. Where can Where can people go to learn more about BMC and Control-M? We've got bmc.com as our homepage, and you can check out demos and tours of use cases.
So, if you're looking at how could you be using smart automation, we've got great ideas there for you. Excellent. Well, folks, that's all for today. Thanks to everyone for listening to E Speaks, and we'll see you next time. Thank you.
This transcript was generated automatically from the
video's captions and may contain errors.
