How This Software Engineer Automated Code Generation and Earned a President's Award

Uh, certainly, there are a few, uh, occasions, and, uh, I would like to, uh, you know, bring you to 2010, uh, 2009, 2010, probably those period of the time, um, definitely in my leadership career as well, I have such situations, but I will again go back to the software engineering world where I was into the coding side of it. So there, there was one project where, uh, I was part of the team. I was an individual contributor. I was a developer in there. And the, the, the project was basically uh for British Telecom, and we were actually automating their workflows of provisioning VPNs to customers. Now this, this, uh, automation was basically using the Oracle's VPN product and if we want to automate it, you know, you need to write lots of code to change the manual task to the automated task. So, you know, there were around, uh, I don't remember exactly, but. There must be some somewhere around 30-35 of the processes which we're supposed to automate, and it was like every 2 months developers do the development, converts those XML files, creates those XSLTs, write the code, and automate it. Uh, suddenly one day I realized that why I can't generate this code, right? It was basically a parsing of XMLs to create the XSMT. So why can't I create a code which generates code? And, uh, you know, I thought of giving it a try. I created a small uh POC around to it, uh, written a code to pass the XML, uh, go to each and every uh layers of XMLs and creating XSLT out of it, which will ultimately automate the, uh, task. And the POC worked very well. And then I automated the entire process and demonstrated it to my group director. And he was quite impressed with that. So he said, you don't focus on the project, you focus on this, this initiative and you complete it. And I completed it in, I guess, uh, I, I just took around less than 1 month, and I created a code which was actually generating a code. And all the manual efforts, you know, developers spending 2 months in, in developing that automation process, generating those exercises and all, we're all automated. And uh. I, I, I, I, I, I actually reduced those all efforts, you know, to none, because the code was getting generated automatically. Ultimately, I ended up receiving a President Award from British Telecom as well for this initiative.

So biggest challenges was from a technology standpoint, uh, see, I got a, I got a buy-in from my group director, right, because, uh, I cannot simply, uh, spend my time in doing this unless I have a buy-in from my senior managers and leadership. So my first objective was to get the buy-in from my, um, senior leaders, and that's the reason I spend extra time to do that POC and I only focused on one task first. And once those, uh, once that pilot was successful, then only I started, uh, doing it. I got a buy-in. Now, my senior leaders actually managed my time, right? So, uh, they were very interested in, uh, doing this because it was a kind of something which even was saving money for customers. So, you know, spending resources, 2 months, uh doing this automation, and if you just want a click of a button, if you're doing it with just simple uh little bit of changes here and there, even business would agree to it. So, challenges were majorly from a technical aspect which I solved by taking help from my peers, from my network, uh, in other company, uh, reading blogs and lots of stuff. But from a time management standpoint, from a, uh, you know, coming, uh, getting myself out from day to day deliveries, I got the buy-in from my senior leaders by showing the successful pilot.

So, when I have, to be frank, when I started, I, I, I, I wanted to try one alternate, uh, which was basically passing the XML creating the XSLT, and, uh, which worked very well. Now, uh, there, there was a scope of improvement in terms of uh what kind of tools that I can, I can actually opt for. But to be honest, uh, the solution which I was building. Uh, it was like, even, even without considering the alternatives that, that, that, that was actually going to solve the bigger problem, uh, which was, you know, uh, resources spending time in writing code. So my focus was to, uh, get things done quickly because, uh, I wanted to save money for business and I wanted to close this uh initiatives as well. So from an alternate standpoint, I definitely from a technology standpoint, I could have used, uh, instead of XSLD I could have used something else. But because it worked at that point in time, it was, it was making sense for me. I just went out with my gut feeling that this is the right thing and I, I implemented it.