What We Covered

In this session, we explored how artificial intelligence is being applied across field operations within AbaData. From inspections and reporting to planning and predictive insights, you’ll see how AI can help teams operate more efficiently and make better decisions in the field.

AI in Field Operations

An overview of how artificial intelligence is being integrated into field workflows, helping teams manage inspections, reporting, and operational data more effectively.

From Complex Workflows to Simplified Reporting

See how traditional reporting structures—often involving multiple report types, parameters, and manual setup—can be streamlined using AI-powered tools within AbaData.

Smarter Planning and Scheduling

Explore how AI can support more efficient routing, smarter task planning, and improved adherence to inspection schedules.

Identifying Patterns and Risks

Learn how AI can uncover patterns in data, highlight non-compliance trends, and support proactive decision-making across your operations.

Beyond Reporting: AI in the Field

A look at emerging applications of AI within AbaData, including predictive production modelling, interpreted video inspections, and photo analysis.

Free to attend

Live demonstration

Recording available after the event

Stephanie Verity
‍“Morning, everyone! Thank you for joining us today for our webinar, AI For Field Operations. My name is Stephanie Verity, I'm the lead of customer success here at AbaData, and I'm going to be your host again for today's session.

We're excited to jump into Part 2 of our series, exploring how AI is being applied within AbaData.

If you've missed part one, don't worry about it. The recording is available to all of our users in our support hub, right within AbaData, and we're going to share a link to that in the chat here in a moment, and instructions on how to watch that are also going to be sent to you after the webinar as well.

So before we get started, just a couple of quick reminders for you. We are going to be recording today's session, and we're going to be sharing a recording with all of the attendees afterwards. Questions can be submitted anytime, and we ask that you use the Q&A panel at the bottom of your screen.

If we don't get to your questions today, we're going to follow up with you directly afterwards, so please feel free to, look into that as well. And as mentioned in our last session, our focus at AbaData has always been on helping our users better access, understand, and act on their data.

We're excited to share more about how we're leveraging AI to transform field operations in today's session.

So today's presenter is Kurtis Poettcker, founder and CTO of AbaData. Under his leadership, AbaData has built one of the most comprehensive energy and software platforms in the world, serving nearly 30,000 users.

With deep expertise in GIS, software development, and energy data systems, Kurtis continues to drive innovation at AbaData, including the integration of AI into the company's next generation of products.

Today, he's going to be walking us through how AI is being applied within Field Ops, and what that means for the future of field operations, inspections, and decision making. And so, Kurtis, I'm going to hand it over to you.”

‍Kurtis Poettcker
‍“Thanks, Stephanie.”

‍Stephanie Verity
‍“For sure.”

‍Kurtis Poettcker
‍“Great job, as usual. I called you and your team out on the first one, and I'll do it again this time. The work that you guys do on support and success is really one of the differentials that sets AbaData apart, and

know from the development team and from everyone at AbaData, we really appreciate all the effort, and I think our users would echo that sentiment as well.

Stephanie Verity
‍Thanks so much, Kurtis.

Kurtis Poettcker
‍Awesome.

Great, so yeah, Stephanie did a great job there, setting the stage. Welcome, everyone, good morning.

This is our second episode in a series of webinars about AI and AbaData.

And this time, we're taking a lot deeper dive specifically down the trail of where AbaData fits inside of field operations. So I'm going to just begin sharing my screen.

And… Hopefully that comes through to everybody.

Hmm, right. AI for Field Operations, which really is about turning field data into decisions.

And as we dive into that.

I think it's important to set the stage a little bit and help understand why this matters.

Throughout Western Canada, oil and gas specifically.

mountains of field data is being collected on a daily basis. But we believe here at AbaData, and I think most people would agree that that data is being underutilized currently.

Most companies, either for regulatory purposes.

Operational purposes, maintenance purposes, are collecting troves of data.

On a regular basis, and storing it somewhere in their systems now.

That may be a sort of digital forms application, it may be spreadsheets, it is even still, hard paper.

Where we're writing things down and putting them in filing cabinets, backs of trucks, etc, etc.

this data.

Being collected, being stored, but rarely is it being analyzed deeply to get the full value out of that data.

And that represents a real gap.

We are incredibly data rich.

But due to that lack of analysis, we're actually, very insight-poor.

And again, we believe the value here isn't only in collecting that data, it's… the true value is what you get to do with that data once it's been collected and is sitting in your various data stores.

Which presents an opportunity.

Anytime there's a gap, there is an opportunity to fill it, and so…

We ask ourselves here, you know, what if…

You could ask questions of that data in plain English.

What if you could instantly spot trends and issues that exist in that mountain of data.

and proactively tackle it, you know, before something happens. And again, just echoing back to the first episode in this series.

What if you were able to… Cut that decision-making time down.

From days to minutes.

And really what we're talking about with this opportunity is using AI as an operational analyst inside of your organizations.

So I think that more or less sets the stage for what we wanted to cover today. Hopefully that resounds with some of you. Just a quick agenda here to let everybody know, sort of, where we're going.

I have a bit of an update on some of the discussion that was happening last week, or last episode, sorry. We'll go and just introduce Field Ops. It's a product inside of our portfolio.

This isn't a sales call, it's… it's simply, I think you need to set the stage to understand the tools.

prior to getting into the real demo, which we'll spend a lot of time on today, demoing the actual application of AI inside of that data. Move on to some future enhancements.

in that, are sort of on our roadmap, and where we're going with this AI inside of field operations specifically. And as Stephanie mentioned, we'll leave time for some live questions and answers.

And if you just…

as you think of them, put those comments or those questions in the Q&A section of the webinar. That would be appreciated. We'll do our best to answer them.

Alright, so some updates since the last episode.

I think everybody who was there would have remembered that we actually ended up overloading the system

We anticipated some participation in the webinar, but admittedly, not quite the level of participation that we received, which made the system a little bit unresponsive.

I've been told by the team that they are ready for that challenge, and are encouraging everybody here to participate and follow along inside the AI Assistant as we go through the different demos, and let's see if we can, if we can

Withstand the onslaught, this time, as opposed to, to what happened last time.

We also mentioned some limitations inside the AI Assistant last episode.

Around the… specifically around the spatial capabilities.

The team, Matt and Kallin, have been really hard at work enhancing that spatial capability inside of the AI Assistant, and you'll notice if you are using it now, it is much more…

Spatially aware of the type of operations and questions that you're posing it.

has a much larger GIS backend to support its analysis and the queries that go along with it. I just got a message yesterday, another,

feature that we discussed last week is coming very soon, and that's…

giving AI the contextual awareness of your current map view. So, you'll now be able to ask questions like, what pipelines are on my map currently, and it will understand what you're looking at on the screen, which is massive.

And of course, one of the most popular messages from the webinar last episode, at least in our office, is that we have now

loaded a fantastic banana bread recipe inside the AI Assistant, and so if you were curious and wanted to ask that question one more time, you should get a good result.

Moving on into what we call here at AbaData Field Ops.

And as I mentioned before, Field Ops is a product inside of our suite of tools.

that plugs directly into AbaData, and makes use of a lot of its functionality and data.

It is a mobile and web platform, so it takes the power of AbaData, puts it into a mobile format on your devices.

It's used for collecting, managing, and analyzing field data, and then, of course, reporting out those metrics and KPIs that come from that.

Importantly, it works offline, so it has full offline capability baked in right from the start.

And I think the most important part of this slide is the last point there. It has native integration with AbaData, and you'll see, as we get into the AI part, where that becomes vitally important in not just isolating your field operations data in a silo, but exposing it to

As many data sources as possible, and we'll see examples of that, of that later on.

What kind of data is currently being, tracked inside of Field Ops?

Well, very simply, inspections, think tank inspections, well site inspections, cathodic rectifier inspections.

Maintenance records, you know,

Oil changes, running hours, those, those type of, of data points.

incidents, so… Hazard identifications, safety and incidents,

Safety forms, safety inspections, all of that lives in there as well, along with production and operational data, so…

I think, well production, or…

Compressor logging, or plant site logs, all of that type of data, all tracked inside of the tool.

Why this is important?

Field Ops is, just by design, capable of capturing structured data and storing it in a consistent format in a very central location.

Which is obviously the foundation

or reporting, compliance, and preventative maintenance, but also, very importantly, the foundation for really good AI analysis. We talked about it last time.

Garbage in, garbage data, garbage results. And so you need to control all three of those aspects of that triangle in order to get, really good data, so…

You know, having… Data collected in a smart manner is essential to that,

the value coming from that data later on. And I'm just going to quickly…

share my phone. I'll just start really quickly with the mobile app.

To give you an idea, like, what Field Ops looks like.

At the… at the mobile side.

We should see it showing up here shortly. Sorry about that.

As I switch over, I'll get rid of that.

Field Ops is an application that's customized to each of our customers, so…

Each Field Ops mobile application is custom-built, which means it can be custom-branded, custom colored, really make it look and feel like your own application, which it really is.

We have a sample one here for a company that we use internally, and it's just called Tranquility Resources. You see the app there on the middle of the screen. I touch it to open it up.

The application opens, and you're taken to a…

what we call status page, inside of the application, so…

This is a more or less a report card. Of all the tasks being currently done inside of the Field Ops application, what are their statuses of the… of the

of the inventory that I am concerned with. So right now.

I'm an operator on a run, I have my runs loaded, and I'm able to visualize the status of, for instance, my well sites. I can see I have 180 well sites.

Some of them are blue, that means they're due for inspection, some are green, that means they're completed, and some of them are red, which means they're overdue. And if I wanted details, I can obviously expand into there.

where I'm able to look at those, those pieces of information.

And it goes across all of the modules, so as you go down, you get a breakdown of what's happening.

We're looking here, maybe.

I've been tasked with getting my annual well site inspections up to date, so I want to start to visualize what's currently overdue.

I touch on the overdue status, and I'm taken to a list of all of those well sites that are currently overdue for inspection.

I can see…

As I scroll down the list, I have a bunch of items. They're sorted by the distance they are away from me, so closest ones are at the top, that type of an idea. I can touch into them, I can see their current status, some information about them, as well as their inspection history. So I have access to

All of those inspections that have occurred on this particular well site.

My job here is to create an inspection, so I enter the inspection form, fill out the questions as I go through.

And you'll see there's some advanced functionality that happens inside those questions, triggering things like noncompliances, deficiencies.

attachments, all of those, those type of typical operations. If I just back out of there…

I'm also able to access an AbaData style map right on my device, again, in a disconnected status, so…

I can see us here in Crowfoot, in northwest Calgary. That's where we're currently located. If I zoom out to where all of my inventory is located, this is all of those overdue well site inspections, so it's respecting that same filter that I set

I can zoom in on the map.

I can see that we have some pipelines, some wells, as I touch on each of these objects, just like in AbaData.

I have a flyover, and I can touch into them to access more detailed information about those particular assets.

It's a full-function map, so I can add, for instance, satellite imagery if I was interested.

In… in taking a look at what that area looks like, you have, again.

A lot of mapping capability that is built into the application.

That's really a very quick run-through, but again, I just wanted to sort of set the stage and give you an idea of what that mobile aspect of Field Ops is all about.

I'll just quickly switch right back to…

my computer, and show you what that looks like in AbaData.

There we go.

Yeah, shame.

So in AbaData, when you have Field Ops enabled, you get access to… you'll see a little bar at the bottom.

Which is your portal and your entry into the field ops, itself.

But there's more than that. There's also tie-ins with all of your map objects. So, if I were to look at this well, double-click on it, no different than a regular AbaData double-click.

It opens an information page. However, you'll see there's a couple of new tabs now. So, we have this one called Field Ops Management. This is about managing your inventory on a site.

And so you're able to, for instance, shut this site in, where you could set everything to inactive, and now all of those items are inactive.

There's also a Field Ops Inspections tab. This gives you access to…

that history of inspections that have occurred on the different pieces of inventory that are on this site. So.

We've got well sites, we've got some presco switches on here, there's obviously a pig run that starts or ends here.

And some tanks on site. And you can access all of the inspections relative to that particular,

inventory that's on that site. Again, we've got 5 pages worth here, so you can see we can go back, all the way back to 2022, if we're interested in that.

Same information is available for pipelines, so if you double-click.

You can, again, get access to that core AbaData functionality.

But also look at pigging records. So, you know, if you wanted to see all of the pigging events that had occurred on this pipeline, it's now available right inside of that pipeline double-click.

Expanding that portal at the bottom, this is where you get access to the sort of KPIs and the metrics and the management of all of the components inside of Field Ops. So, you can see we've got some areas set here.

We're currently, as in these areas, we're at about 63% complete, 5% overdue, and 31% pending. And you can break down each of those modules individually to see the metrics on all of those.

Up at the top, you can do reporting. There's a usage graph that'll show you, of all your users, what their, what their usage is looking like.

There's configuration ability, deficiency tracking, so if you were

capturing deficiencies on those inspections. You can visualize them here.

There's ad hoc forms, there's a working alone module that will… can be used for visualizing your workers on a map, so if you wanted to see sort of where everybody was located, you're able to visualize those right on top of AbaData. There's a tasking slash work order system.

AI assistant, which is where we'll get to, I promise, in just a second.

As well as, Details on those individual module… er, inventories, and what their current statuses are.

You can make edits to them, you can create new ones. Again, you can open them to see inspection histories. If you wanted to take a look at a pig run, for instance, you can visualize a timeline of photos that have occurred on that particular

Pig run, and you can see all of the different photos, and the number of days that were taken between each photo. So, it gives you a bit of an idea there, in this case, of potential debris buildup as pigging intervals are maybe set too far apart or too close together.

And each module has the same, really, type of functionality available to it.

So yeah, that, I think sets the stage fairly well for what Field Ops is. Just switching over really quickly to get into the actual demo, why we're all here, and that's to talk about the AI Assistant that is now integrated into that rich Field Ops and AbaData.world.

The AI Assistant, as we mentioned last time, is a natural language interface, very similar to a ChatGPT or a Gemini that several of you are used to using.

It's now, though, built directly on your Field Ops data.

And allows you to explore that data, again, with… without…

Needing to write code, without needing to export data.

Without needing to manage all of this inside of a complicated array of spreadsheets.

What we're talking about, really, is the switch from digital forms to operational intelligence, and

You know, there's a lot of digital form tools out there. They're great at capturing data, great at storing data, and great at basic reporting.

But what you as users of that data should expect.

Is the ability to analyze that data instantly.

Identify trends and risks, and support real decisions. And I have a couple of graphics on this slide.

We really feel like all that data that's being collected is… is a mine, a gold mine.

And most… most organizations are just sitting on that gold mine.

Whereas we want to enable you, with a pickaxe, to start getting access to that gold that's sitting inside that mine.

Where, you know, most companies Have the data, but very few of them are actually using it.

And why this matters in oil and gas?

You know, your field ops data is only part of the picture.

The real power comes when you start combining that with other data sets, such as regulatory data sets, environmental data, historical operating conditions.

It really allows AI to be

Exponentially more powerful when it's able to understand the full context of that data, not just the data points itself.

So, enough talking.

Let's see what the AI Assistant looks like when it is exposed into your field ops data.

A slightly new home for the AI Assistant inside of Field Ops. It exists on the left-hand side here.

When this screen is minimized, it, of course.

Still lives in its regular place there, but we wanted it to be accessible when you're purely working in field ops as well.

Same interface as last time, you can start to ask questions. So.

Questions can really be divided out into two categories.

We have what I sort of call data cleanup questions, or data anal… not data analysis, data lookup questions, and then we have really analysis of that data question. So let's drill really deep into it, look for trends, look for opportunities, that type of query. So let's start with a simple one.

Data cleanup, do we have any active…

tanks that are inact… oh, sorry, tanks, sorry… on sites that are inactive. And I see my typing has not improved since the last, since the last webinar, but here we have just a very simple query that we're asking.

And really, we're looking at, are we inspecting tanks?

On inactive sites, so potential… a potential for, you know, elimination of some inspection tasks, if, if the tanks, in fact, aren't,

aren't active and being used, then why inspect them, and why have them count towards the KPIs and the metrics that are being reported out? So the AI assistant's going ahead, doing that analysis, and you'll see it comes back here with a answer.

So based on a review of your data, there is one active tank associated with an inactive site, so…

What that's saying is that we have a tank here on 1206 that's sitting on well 1206, but well 1206 is inactive, so why are we pursuing an active tank inspection on an inactive well? And again, as before, there's an ability to download that report if you needed it.

Yeah.

I have a few pre-canned, again, just to… just to make this a little bit more efficient, rather than have to wait for the responses. So, I thought it would be great to just go through a couple more examples of those.

Those instances where we're talking about

that data cleanup. So here we've got…

A question about the work alone module, and we're just really asking.

What's the average number of working alone shifts, over the last month?

And we have an analysis here that's saying we average 20 shifts per day. Here are the summary of that. There's a bit of a bar graph. In this case, we're dealing with sample data, so it maybe is not necessarily reflective of reality, but that's the data in the system.

A table at the bottom.

I want to dig a little bit deeper.

How about the average combined distance travelled per day over the last month? So now, looking at the, you know, the operators that are out in the field using the workload module, how far are they travelling?

average per day. And so here we've got average daily distance. Across the 30-day window, it changes. Peak single day, again, I didn't specify any of these queries, these are just…

sort of interpretations or inferences based on the question that I had asked, what data I might be looking for.

Another bit of a graph here, this time it's a little bit more interesting, and a table as well.

If I wanted to ask a very straightforward, how many total working alone sessions have there been?

Over… overall time? Well, it's just a very simple answer, 2291.

Here we have,

a bit of a deeper dive. I now want a report showing all my sessions, including the user, session start and end date and time, and the distance travelled.

And so, as we go into here, we have a report showing that information, and I'm able to download that as well into a spreadsheet.

it's a little bit of a more interesting query.

We're looking for wells that appear on the AER's D13 list, which obviously is loaded into AbaData on a daily basis.

In my operating area, which is, in this case, Coaldale.

how many wells are on… are on that list? So, I have a list here. It knows my operating area, it infers that, queries that. D13 brings me a list of resultant wells.

I want to dig a little bit deeper, because I'm interested in my Field Ops inspections, so I ask.

How many of these wells are scheduled for a D13 inspection in Field Ops? And it comes back that only one of those wells is currently scheduled for an annual inspection.

Which may prompt me to take some action and schedule some more, some more inspections.

Here's another one relative to well production. So, how many,

Wells have recorded production in the last 3 months.

I'm sorry, have…

Yeah, recorded production in the last 3 months, but have a well, or have a tank inspection

where the status is shut in. So, here we're looking for inspections that may not be accurate, or accurately answered. If someone has inspected a tank and said it's shut in, but the well currently is clearly producing, that's a bit of a conflict, and so this is able to report that out, again, for sort of data cleanup and follow-up work.

Another one related to working alone here. Again, just simply asking.

what's a typical start and end time of all of our, employees? And here we have,

a couple of responses to that. So, shifts typically start between 5 to 9, and they typically end between 4 to 7. And there's some graphing here on

all of the… Number of sessions that would fit under each of those start and end times.

Now I want that grouped by employee, so I drill a little bit deeper, and I have a chart here that will show me their average start and end points per… or start and end times per employee.

More on the data side, so…

How many pipelines are owned by my company, but are not currently being pigged?

It's going to do an analysis, show all of those lines that are licensed to my organization.

And then report out the ones that are not currently being scheduled inside Field Ops for a PIG run to be run on it.

And again, you can download that.

Really, that covers sort of the straightforward, high-level, simple data cleanup. Now let's get a little bit deeper into analysis.

And here I'm asking a question of all the forms that have been submitted

On all of my tanks,

With the most recent inspection, what's the breakdown of active.

out of service and shut in. So that's a question on my form. I'm able to, ask

the AI assistant questions about data that's being captured. It breaks it down. We have some graphs representing that.

But now… I pose a really interesting question, I think.

speaking of that form, and comparing the questions that are on that form with the requirements of Directive 55,

Are there some questions that we should be adding to our form?

Vice versa, you could also ask, are we asking too many questions? Is this form too burdensome, unnecessarily?

Now, nowhere have I supplied, Directive 55 to… to… in this session,

However, the AI assistant has access to public data. Obviously, Director 55 is a public document, and so it's able to analyze the questions that are asked, and

the… The reason behind asking them…

analyze that and stack it up beside the requirements of Directive 55, and identify some gaps. So here, you know, we're not necessarily asking our secondary containment valves locked and closed. You'll notice that it's…

pulling that reference from Section 12.4.4, and identifies the purpose there. So.

Really insightful way of interpreting the questions in your forms.

against…

Requirements, and coming up with some suggestions for improvement on actual forms that could be, that could be created.

Here's another good example of, sort of, deep analysis, where I say I'm the foreman for the Coaldale area, what should I be focusing on?

And it's gonna gather that data for me. It's, you know, it's identified my first priority should be overdue inspection. So I've got 48 overdue regular inspections.

There's also some non-compliances that have been identified, should probably deal with those next.

I also have some pending tasks in the system that need to be carried out, and so it really just is a bit of a breakdown with no real guidance.

On… on sort of top priorities as far as field ops is concerned. So again, we only have access to the data that is inside of AbaData.

Here's a performance-related question. So, what hierarchy, or what run, or what field area

has the best record for completing inspections on time, and we get into some performance metrics here, and you can also narrow that down by time, so…

What about last month? Is it any different? And we have a bit of a different result to that query. I thought this was an interesting question.

very open-ended.

No guidance, are we getting better or worse?

It's inferring that we're probably talking about field ops data.

And so it comes through and reports out, you know, some very interesting metrics.

I'm also able to drill a little bit deeper there and talk about what's changing over time.

And so as we go through, it's identifying trends in my data, either positive or negative.

And laying those out in a very easy-to-consume format.

There's some… There's some graphing here as well that is available.

Here we're getting at, battery logs. So, I'm looking at analyzing, the DHI 509 battery.

And here we have, the analysis from that with some average, minimum, maximum values.

And some key findings about that data, as well as, again, a bit of a graph there.

Here we've got a good question.

Do we have any problem areas that have been identified in our veg management inspections, and what can we do to address the risk?

So we've got, you know, heavy presence of broadleaf in Coaldale Southwest, high foxtail and kochia in the northwest, high kochia and Spurge in the northwest. Again.

These are just questions, or answers to questions on the veg management inspection form, laid out in a really easy-to-consume and easy-to-interpret format. We've got some recommendations here, so…

me not being necessarily a veg management expert, I'm able to read this and maybe follow up on some of these recommendations.

If I drill a little bit deeper here.

Are there any mismatches between chemicals applied and the targeted weed species? So, again, AI Assistant.

inherently doesn't… is not a weed expert, but has access to public data, so it's able to analyze the questions on the form that were asked, and one of them is, what weed species are you targeting? The other one is what chemicals and at what rate were applied.

It combines those two questions and comes up with a couple of mismatches here, where we're applying broadleaf herbicide on grasses, and we're applying a grass herbicide on a broadleaf. So.

Again, me not knowing anything about chemical, I'm able to come in here and ask some questions and get some answers that I'm able to understand and interpret.

A bit of a fun one here, just looking for speeders. You know, what's the average speed of our employees? And I can see.

Norman Stone has a fairly heavy foot and, you know, has clocked 146.2 kilometers an hour, which may be something I want to bring up with Norman.

Later on when I meet with him.

Last one here, and then we'll move to questions.

How close to schedule is this particular pipeline being pigged?

So…

Here, we're looking for the pigging schedule for 33143-8. We're coming back that it's a monthly pig. It's identified it's currently 19 days overdue. That wasn't what I asked. I could drill in a little bit deeper.

And I can see that the recent pigging history is below, as well as who did it.

But really what I'm after is what's that average time between pigging for the entire history of this pipeline?

Rather than having to do all that math myself, I just simply ask the question.

And it comes back, the average time between pigging events for that pipeline, 30.3 days. So pretty, pretty close to what, what we're looking for. And obviously, if you're a corrosion expert, you're thinking more along the lines of what's the longest it's gone without a pig?

And those questions, you know, easily asked and answered, by the, by the AI assistant.

Back to… just really finish up on Roadmap.

Again, the point of those… going through those questions wasn't necessarily to drill on the specifics of the questions themselves, but more to outline the capabilities and start you thinking about the possibilities of what can happen when you ask questions

of your AI assistant, and where this is going…

inside of AbaData is to enhance that knowledge base, rather than relying purely on operational data and public data, to start supplementing it with your internal operational data.

Standard operating procedures, engineering guidelines, optimization playbooks.

If the AI assistant is able to access that information.

What we're able to do is combine it and start asking questions

Not spec… not generally how are we doing, but very specifically, according to our internal policies, how are we doing in our field operations?

It's a real opportunity to take that AI assistant.

And have it start to think

More like a field engineer, as opposed to a data analyst.

A little graphic here that just, I think, represents it, quite well. We're feeding in Field Ops data from the side, operational knowledge from this side, all into the AI assistant. Your query goes into that assistant.

churns through what it knows and really spits out, you know, context-aware decisions, or insights, allowing for better decisions and faster analysis is what we're after. So that's, near-term on the roadmap.

A little bit further down the road, turning that AI agent into an active field ops agent.

Where we're actively running processes in the background without waiting for prompts from specific users. So, think about, asking the AI assistant, send me a…

update of our metrics in Field Ops on a daily basis. And… and then boom, you would have a report sent to you on a daily basis, just, with that simple text command.

Also, things like proactively alarming on trends. So, identifying those trends and proactively responding to them, as opposed to, again, waiting for a prompt to come in and ask about areas of concern. That's what we're talking about, and that's taking that AI assistant

And making it start to act like a field engineer.

Not just think like an engineer, so…

a bit of a distinction there, for sure. So, that's our quick roadmap. We have about 5 minutes left, it looks like, for some questions.

I will turn it over to Stephanie.

And see what we've got in store.”

‍Stephanie Verity
‍“For sure, thanks, Kurtis. And just a heads up, we've had a power outage here at our Red Deer office, so if I tend to drop out immediately at some point during this Q&A, I apologize, and Kurtis, I'll have to leave it up to you.

But we have had some great questions come in so far during the webinar. So, the first one we have, which LLM model powers this AI assistant, and how do you ensure that data stays secure?”

‍Kurtis Poettcker
‍“Great question.

First point, I think, to talk about there when we talk about security.

is that all of this Field Ops data is opt-in only, so we do not do anything with your data until you ask us to.

We do have a incredibly talented, Data science team here.

Who keeps up with the latest models.

We use several of them.

We carefully review each model's security policies to ensure the privacy, protection, and security of all of our customers' data, and we'd be happy to have those conversations with you individually to discuss the specifics around that, for sure.”

‍Stephanie Verity
‍“Okay, great. Another question that came in, will AbaData AI pick up which sites may require a D13 inspection based on what AER has for data, and can we update manually and submit as needed?”

‍Kurtis Poettcker
‍“Yeah, great question. Absolutely is the answer. That's the easy, fast answer. We're getting the D13 data from the AER on a daily basis. You can access it today inside of AbaData. So absolutely, that,

that will be picked up and reported out by the AI assistant at the time of query, for sure.”

‍Stephanie Verity
‍“Awesome! Well, that's all the time that we have today, so thank you everyone for joining us, and for all of the great questions that you submitted. Any questions that we didn't get to today during the session, we will follow up with you, make sure that we get those answers to you as well.

If you're interested in learning more about Field Ops, and you're interested in one of our team members reaching out, you're gonna see a poll that's gonna pop up on your screen here, and you can just indicate that by answering the poll.

So, all attendees are going to be receiving a follow-up email with a recording of today's session, and that's also going to include those instructions on how you can access Part 1 of the webinar within our Support Hub in AbaData.

If you have any further questions about AI at AbaData or any of our other products, please feel free to reach out, we are more than happy to help. And more information on how to contact us can be, found in the resources tab here in the Zoom.

We look forward to having you join us again for future webinars here at AbaData, and we're going to be exploring the latest innovations within the company. So thanks again, everyone, and have a great day!”

‍Kurtis Poettcker
‍“Thank you.”

AI For Field Operations

Exploring how AI is transforming field operations, inspections, and decision-making within AbaData.

What We Covered

AI in Field Operations

From Complex Workflows to Simplified Reporting

Smarter Planning and Scheduling

Identifying Patterns and Risks

Beyond Reporting: AI in the Field

About the Speaker

Kurtis Poettcker, MSc

Discover the future of energy operations