Codesmith’s expansion of the software engineering immersive program to encompass AI and ML reflects the tech industry’s growing appetite for engineers versed in these tools.
But Codesmith residents are already securing modern engineering jobs in AI and ML by integrating these new tools into their open source projects (OSP) while on the program, graduating with a strong case for tech hiring managers.
Jordan Palmer, who graduated in January 2024 and now works as an AI Technical Product Manager for Riverside Insights, built an AI-powered tool to simplify the use of Kubernetes for other software engineers.
Originally from Jamaica, Jordan moved to the USA to study business, but a stint in mergers and acquisitions after college left him unimpressed with the field and more intrigued about the AI tools he had been exposed to on the job.
“I was hired in May four months after graduating. Riverside Insights is an Ed Tech company where I build AI applications. As a technical product manager I'm actually building and coding out the products.”
His curiosity in tech was driven by the explosion of ChatGPT in 2022 and a realization of the myriad of possibilities AI was going to bring to different industries. While he hadn’t previously worked in tech before Codesmith, after a month of intensive studying using Codesmith’s free online learning platform CSX, Jordan passed the admission test.
“I actually did business in college and started working in mergers and acquisitions, but I wasn't a fan of it. I decided I wanted to change industries and learn how to program, learn how to code. That's how I ended up at Codesmith.”
Jordan lives in San Francisco surrounded by software engineers. When he decided he wanted to learn how to code and apply to Codesmith “they were like ‘you should go for it.’ After two months of studying, he was admitted into the immersive program.
“Initially I wanted to be a software engineer, but I also knew that I wanted to learn how to integrate AI into the apps I build.”
Knowing this, he made sure that in every single one of his projects at Codesmith he integrated AI. The aim was to put him in good stead for an AI-based role after graduation and his final OSP, KubeVX, did just that.
“I spoke about the apps I built at Codesmith a ton in my interview and the interviewers asked me so many questions about my OSP as well, because what I built at Codesmith was similar to the apps I’m building now. It’s leveraging AI to take data that’s opaque and making it actionable”
The inspiration came during the program when Jordan was introduced to Kubernetes and saw how AI could simplify the tool for software engineers who are increasingly asked to work with the software.
“When you have a large-scale application,” Jordan explains, “the way you deploy it is almost in a decentralized manner with different apps running simultaneously. They're not all connected to each other, meaning, they're not all talking to each other.”
Technically, these are micro apps and are deployed using containers. Kubernetes is the software used to monitor the resource consumption and scale of these containers.
Jordan and his team saw how the changing nature in how modern apps are deployed—from centralized to decentralized structures—has made Kubernetes a necessary tool for engineers.
“But the issue with Kubernetes is it's really technical. Software engineers often aren’t familiar with it as it's more in line with DevOps. But as many companies are transitioning to this architecture, their software engineers need to use Kubernetes.”
His team saw this as a perfect use case for AI. “AI understands Kubernetes in a way an individual may not. So we thought: “What if we fed AI the Kubernetes information?”
Now users who need to interact with Kubernetes don’t need to look through different charts and graphs, instead KubeVX delivers the data in plain language.
Building this product gave Jordan such a strong foundation for AI job interviews that he landed a role just four months after Codesmith.
“Use products you build during the program to showcase your AI skills and understanding”
Riverside Insights is owned by private equity firm Alpine Investors Group, who began an initiative to implement AI at their portfolio companies.
“Riverside Insights is part of a group of companies exploring AI-driven innovations, and I was brought on to help demonstrate the potential of AI in education technology,” Jordan says.
He was one of four engineers chosen by Alpine, landing at Riverside Insights with the task of demonstrating AI’s potential to help the company.
“It was great. I built a bunch of stuff for them and they brought me on full time.”
AI is an expanding field, but Jordan explains that it actually requires less technical expertise than many people think.
“You're often essentially taking the AI model and embedding it where you want by using the API. And so many companies want to do this right now because the use cases for AI are endless. Both internally, to aid employees, and externally, embedded in products to improve user experience.”
He describes how his job interview largely focused on the apps he built at Codesmith and KubeVX was central to him landing the job.
“KubeVX really carried me through my interviews. My choice to focus on building AI products at Codesmith was central to me landing this job, I think about that all the time.
“I spoke about the apps I built at Codesmith a ton in my interview and the interviewers asked me so many questions about my OSP as well, because what I built at Codesmith was similar to the apps I’m building now. It’s leveraging AI to take data that’s opaque and making it actionable.”
Riverside is an EdTech company focused on testing solutions and platforms to help teachers gather insights on students.
While Jordan is unable to share details about specific internal processes he states that most of the AI use cases Jordan addresses when building products involve “leveraging AI to improve educational tools and provide actionable insights for users, focusing on enhancing the experience for both educators and parents.”
“If I were to talk to current Codesmith residents I would tell them to build AI tools"
Jordan says the best way to start is by familiarizing yourself with the APIs for the frontier models like GPT-4o, Claude, etc.
“And it's a lot easier than you think. You don't need to be a machine learning expert to do it, you just need to read the documentation. That allows you to build functionality of these models into any apps you're creating.
Building useful apps so you have something to show interviewers and talk through, as Jordan did with KubeVX, is something else he emphasizes.
“Use products you build during the program to showcase your AI skills and understanding.”
The team environment during the OSP portion of the program provides the perfect opportunity to build something ambitious that will set you apart in the job search process.
“Sure you can build apps alone but you can achieve a lot more in a team. For KubeVX we started by creating a list of features and modules that we were going to build, listed in order of priority.
They then divvyed up the tickets so one or two people worked on the same feature. If it was a particularly big feature one person tackled the frontend and another the backend. But for smaller features just one person worked on it.
“Once we decided who was going to do what, we went our separate ways and just grinded, having periodic check-ins to monitor progress. When needed we'd move people around to address roadblocks (throw more people at tough roadblocks that might have been holding us back). And then when done working on individual features we'd pretty much "hook them up" to each other to create a finished product.”
He says that the biggest obstacle was also the reason they decided to build the product - understanding Kubernetes, “it was extremely difficult to understand how it worked.”
Jordan acknowledges that the hiring market for traditional software engineers is challenging today, but he is also seeing enormous possibilities open up around AI, and wants to encourage other residents to take advantage of these opportunities.
“If I were to talk to current Codesmith residents I would tell them to build AI tools.”