CTO's Perspective _ Unlocking the potential of generative AI ✨

We've had a lot of really exciting announcements here at automation anywhere over the last couple of weeks .

And I'm super excited to have the time to be able to sit down with our Chief Technology Officer , Prince Coley .

Thanks Mike .

I very happy to be here .

So the first thing I want to ask is of all of the stuff that we've announced recently .

What are the most exciting things that you're looking at launching and how do you see them improving our customer experiences ?

Awesome .

I love that .

That's the first question .

So there is uh you know , we are the way we are approaching generative A I um everything about our product and our customer and how they interact with this will change .

But I will talk about 22 particular ones which I think will have the most impact .

Uh One is um the most common .

There are two most common set of people who interact with our platform .

They are what we call the bot creators autom and then the people who are the users of the bots , the automates their life will change because instead of having to really either follow a user and tell the user to do what it is that they do or to in some ways program a bot , they can simply tell the platform , they can say make me a bot that takes data from this email that gets this stuff out , sends it to this particular , you know , in or or sales force and then takes this data from there and then create this dashboard or create this email or create something else and then share it with these people .

It is conversational , it is uh it is makes it very , very accessible .

Um And finally , it also allows you to talk to it in a language that you are most familiar with .

The second piece is uh uh a lot of the data in enterprise is still based on documents , whether it is email attachments , whether it be , you know , just emails that people send , whether it be something you download some web page , whatever it may be .

And while we have been able to handle structured semi structure documents for many , many years , the biggest change with NA I is our ability to interact with unstructured documents will suddenly go through the roof .

And now you can simply say that in this particular contract which may be completely unstructured , just find me these things and do this with them .

You can do analysis of that , you can do action on that , you can write new documents for that .

You can summarize using the right terms , very quickly .

Yeah , I like that .

And I think straight through processing rates will also increase , right ?

Even even for document types that we've had success with previously a structured or a semi structured document .

If we missed a particular field in the way that we've been extracting data previously , we can now use generative A I capabilities .

It's almost like a backstop say like , hey , I missed this field .

Give me this one .

What about automation ?

Anywheres architecture makes these kind of rapid innovations that we're talking about really possible .

So many years ago , we took a step where we went completely cloud native micro services based and container based .

Now the reason we did that obviously was to make it a much more modern stack , be able to go to any cloud , be able to auto scale , be able to have high resilience , right ?

And all those good things .

Um One of the advantages of that of course , is given the molar of the of the platform we are now able to add .

And now for the last few years , we have been able to add new functionality um like a package with the package very , very quickly and we can do that in so quickly that you don't have to wait for a release , we can just release it whenever we are ready , we push it out , customers can download it and use it whatever kind it may be .

So if you take it , look at open a I , when they release , uh we were the first people to write a package to , to integrate it um natively and our community was the first one to start using it , you know , for their own use cases that they had in mind .

And I think that that modularity also speaks to resiliency of your automation , which is interesting , right ?

Like something can be fast but is it also resilient ?

In this case ?

It is right ?

Because not only are you able to push out those new package versions , but if you created a bot six months ago , it's gonna be based on the same package version that you built it on .

And so when it deploys , it's gonna be indifferent , right ?

So I I really like that concept and I think the architecture lends itself well to stability .

I know customer trust is a huge thing for you .

Stability , but also innovation at the same time and governance because the enterprise governance is so important , right ?

They should be able to say , what am I using ?

When am I using it ?

And when do I move to the next version of it and you can control all of that ?

So modality really allows you to do all of that as opposed to getting a large monolithic software which you use it as is and if it fails , everything fails .

I'm curious kind of shifting focus here as a CTO .

How do you use generative A I to make your own job , easier , enhanced Prince 2.0 you know , um Every joke I do I tell , I ask GP to regenerate it first .

So usually , you know , that's the case .

But most seriously , um I've been playing around with it now almost since the first day .

Um And obviously it has gotten better , a lot better since then , but most of the time I spent today with it is trying to get it to use like a customer word , right ?

What can it do and what new emerging capabilities it it may have that allows someone to interact with it in a way that is different , unusual but in a useful way .

And we , you know , we we get , we see new things all the time , especially for example , for the unstructured documentation part , we were able to really just by playing with it , we were able to extract information from documents that customers find very useful .

And that is how can the gene that was the genesis of how to integrate it with our platform to extract the data as well ?

Yeah , switching gears a little bit , you know , on Twitter , on TLDR , things like that .

You see that every week there are new generative A I capabilities tools platforms .

There are new libraries available for Java and for Python , there are new um javascript frameworks .

How do you make a determination as a CTO of like , hey , if it was me , I want to build on everything Right .

Like , oh , that looks cool .

Let's integrate it .

But then , like , that product dies three weeks later and we're in trouble .

So , how do you make those determinations to say this is where we're gonna place our bets ?

Yeah .

And you know , it's a great question because that's something that I think every R and D team , uh , especially software , R and D team faces pretty much every day .

Uh , and there are new things , wonderful , new things that come out .

Uh But like any of the investment , one makes one looks at a new set of libraries or a new way of doing things and says first , is this long term Bible or is it just fun and games , fun and games is fine .

But if you really want to kind of be your company injuring foods on a new stack , um you have to look at is this a long term advantage to you and your customers ?

The advantage can be in ease of programming , lower bug rate could be ease of trouble shooting , right ?

You know , there are wonderful languages out there that stop you from shooting yourself in the foot via memory via memory leak .

For example , you know , you should use them .

Um and then um so long term viability , ease of use , ease of migration , you know , you have a millions of lines of code .

Um What should you be doing ?

Should you be taking those and then moving forward ?

Should you be writing new modules in that a new framework .

And if you're going to do that , how do the frameworks interact ?

And then does that change the size of your binaries size of your packages ?

You have to look at all those things in general .

Our engineering force wants to use new , new things very , very quickly uh which are encouraged because it lets you learn things as well .

Um And then , you know , you , we have an architect forum where we talk about some of these new things and over there , we choose what makes sense to invest further in and then , you know , you go slow and study .

Yeah , that's really interesting .

And there's a lot of considerations .

I mean , obviously the newest and shiny thing is always really fun to try out , but there's also considerations for what does this cost me to execute ?

Right .

So when I'm a fully cloud native platform , I have to worry about CPU cycles and , and what is this gonna mean for how large my deployments are and things like that ?

So those are all really interesting and you're right .

I mean , as you said , it's , it's a a lot of dimensions to that problem .

Uh But that's true for engineering , I think , you know , you , you tend you like heart problem and you look at it from all sides and say , OK , that makes sense .

Yeah .

Yeah , I wanna look forward now .

It's 2030 .

What impact has automation anywhere made on the world and specifically on workers lives and uh the future of work .

You know , I , I want people when they go into the office to be happy about going to the office .

I think one of the biggest complaints about enterprise work has been , it is repetitive , it is a mechanical , it is robotic and it , and they mean robotic and kind of , you know , it sucks humanity out as opposed to it being fun .

I think if we by 2030 my , my fervent and my optimistic hope is people would like to go to work because they have , you know , a Sally or a Harry in their pocket who helps them do the work that they need to do , right ?

Who are able to , they're able to go and say , ok , Harry , you can help me with this most of this boring stuff and you will do it with high quality and on time and I can spend time improving things , thinking of new ideas , working with my colleagues with my customers directly because that level of human in intent and empathy , you know , no one has time for right now .

Um So I I very much hope things will change in , in that direction .

Awesome .

Thank you so much for sitting down and having a conversation .

I really enjoy chances like this for us to geek out and talk about our platform and talk about where things are going .

So I really appreciate it .

Thank you ?

Cool .

Thanks everyone for joining us .

Catch us next time .