James Gorman

James Gorman, I'm the senior director of data and engineering at RigUp. I've been working at RigUp for basically the past eight years. We work in oil and gas, basically helping companies find service companies or workers to help them with their projects. They need to get a project done, we help them connect the right people to the right jobs. And then I work on the platform and our data infrastructure that helps make all that happen.

Yeah, it's probably a good way to put it. It's actually interesting in that we get to help with a lot of different phases of projects. Whether companies are looking to drill new sites, repair or do maintenance, or let's say go cap abandoned wells, we can basically help companies go find those services or people that maybe they're not used to having, or they want to have more flexibility in their workforce. We help them find those, put them on there, and then keep them happy and motivated.

Mostly domestic in the lower 48, although we are recently expanding out into Alaska and Canada as well, and then looking to other international countries as well.

Both. There's a worker aspect of this and a client aspect of this. Those workers get to choose which companies they're going to be subcontracting through, and in that case the workers would be primarily choosing us. But also a lot of times we'll go work with all the major operators, and they'll want to staff maybe 50 or 100 people on a given set of projects, and we'll help them as well. So it really is a two-sided thing. When we were starting up, we were more worker-focused; now we definitely focus on both. They're both our customers.

Yeah, we have. It's fun to go out there. It's definitely impressive what they do. It's hard work. Being out there in the field, a lot of time in pretty remote places, puts that in perspective. We work in an office, and that's a privilege in terms of what we're able to do. But it's really impressive to see what these guys are out there doing.

I've worked with Howdy a little over five years now. We're coming up on six from when we hired our first developers through Howdy.

It's been amazing. It really has always been a partnership, and from the different phases that we've had — from our needs changing in terms of what types of engineers or people we need, working in different roles. Originally we hired on a couple of back-end engineers from there, went front-end, full-stack, different technologies. We've hired data roles. Our operations teams have worked with Howdy as well, hiring invoicing operations or payroll operations positions. So it's really been wonderful.

The culture fit has been amazing. Our organization is a mix of remote folks in the US. We have an office in Belfast as well, in Northern Ireland. And then we have people all over South America as well. It really is a wonderful blend.

Yeah, we also use Howdy for AI and machine learning. We actually had our first AI/ML engineer hired through Howdy — Santiago. He's great. I've been working with him for a little over two years now. That's been really critical in us developing models, putting that in our applications, both in our customer-facing apps but also internally as well.

I'd say this is true both for our data teams and our engineering teams or software engineering teams: we're really looking at more generalist profiles. I think that's one of the amazing things about AI — it's really concepts over syntax. What I mean by that is, I need to know what I'm trying to accomplish, I need to know how I want to accomplish it, but the actual details of how do I write this individual code, or what are the individual quirks of the technology, are easier to come by because of AI. That has changed the developer profile we're looking for to more generalist.

I've kind of gone through waves of specialization versus generalist, and there's pros and cons to those different approaches, but what we're finding right now is we want to move more and more generalist. That means our AI/ML engineers focusing a little bit more on the software stack, and learning a little bit of front-end and back-end and how can I actually code in the areas that my models are plugging into, and vice versa — how are our software engineers, or even on the data side, leveraging AI models and maybe doing a little bit of model development as well, understanding those concepts. To me, it's all blending. The generalist profile is extremely important for us.

Yeah, I think that's definitely true. For us, a big thing we always say culturally is, we want people who are going to own the result. My job isn't done if my code is good or if I did my part of the process — it's, what's the result? Did we achieve it? Having people that get obsessed with that and are willing to contribute to any part to achieve that result is key for us. We're finding that a lot with our team.

Howdy definitely enables us to shift our profiles, the people that we're bringing in. A lot of that, as we're moving more from a specialized stack to a generalist stack, that's changed the way we've hired as well. As I'm working with Howdy to bring in new folks, we can be flexible in terms of: what are we looking for? Who are we looking for? Is this the right pool of people? That's certainly helped us to be able to adapt over time.

Autonomous agents, I think, is what we talk about a lot right now. As the agents are getting better and better and they're starting to work in loops, the things that they're able to accomplish are a lot more impressive. We all see that in terms of what we're interacting with — your ChatGPTs or your Anthropics — as you're having these conversations.

The other part that's interesting is the reasoning that's being applied as part of that. As the tool usage is getting better and you're able to plug into different systems, how is that going to change our interfaces? How is that going to change the way that we work or interact? One of the things we think and talk about: we're building an application and a platform — I build a web app, somebody logs in, they use it — but we're also building a chat interface or an agent that sits on top of that and interfaces with it.

Eventually, will it be that we just put out APIs? Will it just be like, "Hey, here's your login for rigup.com — there isn't actually a webpage, you just plug that into your local agent"? How's that going to change the way we interact with things? A lot of what we've built and think about are somewhat constrained by the interfaces that we've built to interact with technology. As those interfaces change, how is that going to change what we build and what we consider a product? I think that'll evolve a lot over time. We're starting to see some of that, especially for developers probably the most, but that'll start applying for different groups and people over time.

Yes.

Obviously there's the Claude bot stuff, which has been interesting to watch people build those up. You have this autonomous thing running on your computer, nudging you on the shoulder and saying, "Hey, I've got something to tell you." But to me, that's not as scary as exciting. It's scary from the perspective of, there's a lot of pitfalls that can come from those approaches and we have to be mindful of that.

The other part that's scary in this: if you think of a common criminal, intelligence may not be as highly correlated to being a common criminal. Now we're giving people extremely cheap PhD-level intelligence. So the baseline of intelligence of everybody in the world is rising. How does that impact us? Those are some of the questions we're going to have to struggle with.

I'm very confident in humans' abilities to adapt. I always say we're cockroaches — we've been around for a long time through a lot of different tough times. I think we will get through whatever challenges come ahead. We're extremely adaptable people. So generally, I'd color myself an optimist.

What makes me an optimist — especially when you think about this from the lens of software engineering — is that we've gone through phases and phases of working at higher levels of abstraction. Whenever I was first studying computer science, I was doing a lot more low-level programming, closer to the hardware, closer to the CPU, driver-level software. As those problems become less on the forefront and you work with higher-level languages, there are sets of things that I don't have to focus on anymore. I'm not doing addition or multiplication or any type of additive processes — I'm using a function that abstracts all that away from me.

I think we're getting that in everything right now. And we've adapted as software engineers. The types of jobs change. What we work on has changed. That's been consistent throughout all of our careers. History-wise, not that long compared to humanity, but even within the last 50 years, that's changed a lot. We're getting that in a lot of different things. People are trying to figure out: how do I use different tools? How do I abstract myself to a higher-level problem? There are things that I don't have to worry about anymore — how do I adapt?

So when I say I'm optimistic, I'm optimistic about people adapting. Adapting is not easy, and for some areas there hasn't been a ton of change — and now we're introducing a lot of change. That makes people feel anxious, or it's hard to imagine. But when you look at different industries — software engineering being one, but there are certainly others — oil and gas is probably a great example: drilling technology has adapted tremendously over the last 50 years, and they've adapted how they do projects and how they work and what a field site looks like. So I think that'll change as well. Humans are extremely good at adapting and generally being happy no matter what circumstance we have. That makes me hopeful.

I love Howdy because from hiring to onboarding to management, I have a partner and they're a Slack away. I know them like I know my own team — both the people internally at Howdy and the engineers and people that I get to work with every day as part of our company and what we're building. They make everything easy. So to me, it's a no-brainer.

To me, the difference with Howdy is, one, having a management advocate on the Howdy side that's working alongside my goals. I'll meet with our Howdy managers, I'll tell them, "These are the projects we're working on, here's the things I'm coaching on with this individual," and they're getting dual-sided coaching, both from me and from Howdy themselves, which is great.

The other part is, it feels 100% integrated to our team. It's not, "Oh, this is a set of contractors that we siphon off work for and that they do independently." It's just, this is a teammate. They're integrated with their team. They're working alongside anybody else that's on our team in the same ways, forms and fashions — on all of our meetings, our calls, stand-ups, thinking about strategically what should we do as a business as well. So to me, the biggest difference is, one, having a pool of people that fit that mold and fit that criteria and want to just plug in and work along with us, and also having that consistent coaching and mentorship from both sides.