In 2017 Shakil Kamran needed to make a career change, and fast. At 30 years old with a baby on the way he was looking for a quick and sure path out of retail and into a more stable and lucrative job in tech. Although he had already enrolled in college for computer science, to expedite his job search Kamran participated in a two-day bootcamp called PepUpTech targeted at under-represented groups. PepUpTech introduced Kamran to Salesforce’s Trailhead, a learning platform designed specifically to train users on Salesforce’s products. Kamran earned his Salesforce Administrator certification on Trailhead, and in 2018—one year after beginning his career transition—he had secured a well-paid job as a Salesforce Sales Consultant.
Since the 2010s, Kamran’s success story has been the dream sold by many a tech skilling program: jump straight into a lucrative software development or IT operations career by attending a bootcamp or obtaining a certification, badge, or similar “micro-credential” (the term preferred by many stakeholders in the for-profit skilling space). But AI has fundamentally turned the skills building industry on its head. Vendors are scrambling to figure out how to adapt to the new AI landscape, while individuals looking to shift careers are desperate to determine how AI will affect early-career hiring.
“Bootcamps will have to evolve”
Coding bootcamps—unaccredited (typically), for-profit schools which instruct students in fundamental software programming skills—are well aware that they need to prepare graduates for the realities of the AI boom. For this reason, the coding bootcamp Code Fellows, which targets beginning learners, integrates AI code assistants into their curriculum to teach students how to leverage these tools. But AI’s impact on the bootcamp market goes beyond the remit of additional lessons. A number of critics note that AI’s threat to the bootcamp industry is existential. Mark Thompson, senior developer relations engineer at Google (more often known by his Twitter/X handle @marktechson), is grappling with this new reality. He explains:
I think that bootcamps will have to evolve … What’s happening with AI, and ChatGPT, and all of these tools that are able to produce code and thoughts and artifacts with such high fidelity, they will remove the lower level jobs.
Historically coding bootcamps have prepared graduates with only a baseline proficiency in software engineering. They rarely teach soft skills like working on teams, or else skills gained from experience like troubleshooting buggy brownfield codebases and grappling with complex Git versioning issues. In order to adapt to an AI world, Thompson argues that bootcamps and certification providers will need to offer more focused and in-depth training to prepare these engineers for more than entry-level work. This will likely require an extensive programmatic revamp intended to set graduates up with the skills and experience typically possessed by mid and senior level professionals. Coding bootcamps must prepare graduates using a more in-depth and robust curriculum in order to bypass the “lack of experience” trap that AI is currently exacerbating.
Prospective learners interested in reskilling into the software development profession are already seeing the fallout of AI’s threat to the bootcamp industry. When I attended the Atlanta-based tech conference Render ATL last year I met a graduate from the Ada Developers Academy, a bootcamp targeted at women of color. From its 2013 founding, Ada has helped students get their foot in the door of real companies by making a five-month internship part of the curriculum. This commitment ensured that Ada’s graduates left the program with a firm understanding of how to contribute to an engineering team, possibly with a job offer in hand. Unfortunately the internship requirement dissolved owing to a lack of companies willing to participate in the program.
Is Chat GPT to blame?
byu/Breezygemeni inAdaDevelopersAcademy
Some Reddit users on the Ada Developers Academy forum suspect ChatGPT is to blame for the lack of internships open to bootcamp graduates. Although others blame Ada’s leadership for poor planning—an accusation that organizational restructuring including the resignation of CEO Lauren Sato, a 45% staff reduction, and corresponding cuts to the student cohort seems to support—there is good reason to blame AI. The sort of boilerplate, low-stakes code that interns and junior developers have previously handled can now be written by not only ChatGPT, but also a range of dedicated AI code assistants including GitHub’s Copilot, Sourcegraph’s Cody, GitLab’s Duo, and Amazon’s CodeWhisperer, among others.
Tech Hiring
AI’s threat to the coding bootcamp industry feels different from prior death-of-the-bootcamp scares that reached their height in 2017 with the shuttering of Dev Bootcamp and Iron Yard. Bootcamps and micro-credentials just aren’t able to deliver a sure-path to a career in the manner they did in previous years. While ChatGPT is already claiming jobs in copywriting and marketing, tech careers appear to be in danger as well. For this reason, although Kamran continues to uphold Trailhead as a fantastic resource for individuals looking to shift into a career in tech, during our recent conversation he agreed that “just having a certification is not good enough anymore.” Following the pandemic, recent tech hiring freezes and layoffs, and the AI/ML boom, transferring skill-based training into a career poses significant challenges.
Unfortunately for junior-level developers—a group which includes new CS graduates from traditional universities, self-taught engineers, and bootcamp grads hoping to reskill into the profession—layoffs in the tech space have flooded the hiring pool with more experienced developers. Hiring managers are bypassing junior developers and shuttering their internship programs in preference for employees and contractors possessing mid and senior level skill sets.
It's hard for me to justify hiring juniors when seniors are delivering so much more bang for buck. Fundamentally, the ratio of senior dev to junior dev salaries needs to go up quite a bit for juniors to be desirable.
— David Hinkle (@Drachs1978) April 28, 2024
Tech hiring managers want to see evidence that potential employees can contribute to complex, messy brownfield projects, but how can engineers who have recently skilled into tech hope to gain this real world experience without first getting their foot in the door? This is a challenge in every industry, and not one from which software engineers have ever been immune. However, opportunities for junior developers and interns were not impossible to find, and after locking down their first position, job security seemed assured. In fact, my developer friends often joked about seeing engineers with 3 years of experience being hired as senior engineers. But the halcyon days in which there were more open positions than qualified individuals to fill them seem to have evaporated. This is a boon to hiring managers, but leaves in a bind both aspiring devs and the institutions (universities, bootcamps, microcredential issuers) hoping to train them. For coding bootcamps that focus exclusively on training up junior developers this is particularly worrying.
Junior developer with 10 years of experience
byu/Familiar_Stage_1692 inProgrammerHumor
The prospect of a downturn in the bootcamp industry will also have potentially negative repercussions for diversity within the tech industry. Let me preface this by stating that bootcamps 1) do not solve all issues around disparities in tech, and 2) can be predatory towards the very groups they purport to serve. That said, these programs have done much toward diversifying hiring pipelines in tech by encouraging underrepresented entrants to the field like women, who “make up 33% of the tech-related workforce,” and minorities. Bootcamps seem to constitute an especially necessary support system for the black community in the US as, according to the NAACP:
While Black people comprise 13 percent of the U.S. population, they represent only 7 percent of the computing workforce
Because so many coding bootcamps have carved out a niche empowering women and people of color to achieve careers as developers, the prospect of losing these resources is disheartening. As career opportunities for early stage coders dissolve, avenues for inclusivity will suffer as a consequence, as well as the corporate diversity initiatives they purport to bulwark: many of which are also under fire, often owing to, as the Harvard Business Review explains: “leaders need[ing] to initiate a dialogue with employees about their true experiences.”
So far I have restricted this post to anecdotes rather than statistical data, and for very good reason. Indeed, I recently commiserated with Cat Hicks, VP of Research Insights at Pluralsight, about the unfortunate lack of concrete data available around the coding bootcamp market. Job placement and salary numbers are especially murky. According to Career Karma, for instance:
Overall, our 2023 State of the Bootcamp Market Report shows that the US bootcamp industry continues to grow, which indicates that students are still finding value in this form of training to start new careers in tech.
While Career Karma’s findings must be approached with skepticism owing to its financial interest in the success of the bootcamp industry, their survey pool of 105 bootcamps is significant enough to warrant attention. Precision Reports echoes this finding, and forecasts that the “Coding Bootcamp Market size is projected to reach Multimillion USD by 2029.” A recent survey by Forbes looks similarly rosy, with 80% of bootcamp graduates self-reporting salary increases. All of this third party analysis suggests that reports of AI’s challenge to the bootcamp market has been greatly exaggerated.
And yet news around coding bootcamp staff retention tells a bleaker story. Layoffs have been rampant this year at popular venues including App Academy, Codeup, Turing, and Tech Elevator, as well as certification providers including Pluralsight, and A Cloud Guru specifically.
The Future of Bootcamps in the Era of AI
The shifts in the skilling space that AI has brought about are just at their beginning stages. However, we can make a few assumptions about where the industry is heading with relative confidence.
One possibly obvious trend that will continue is skills vendors’ acknowledgment of the AI/ML domain’s ripeness for training opportunities. Several bootcamps have sprung up to fill this demand including the AI Bootcamp at UNC-Chapel Hill and Columbia Engineering’s AI Bootcamp. A number of the major providers in the micro-credentialing space including Microsoft, Google, AWS, and Salesforce are also offering training intended to meet the needs of business users unsure about what AI and machine learning signifies for their future. Questions in regards to how AI transforms security, productivity, and data management (in the context of these providers’ proprietary AI products, of course) are well positioned for educating business users.
Less obvious is the path that will enable coding bootcamps to succeed in the coming decade. Even if hiring for developers picks back up to pre-pandemic levels, roles for junior developers and apprentice programs will continue to suffer owing to AI code assistants. Fundamentally, the content and goals of skilling programs will need to adapt in order to reconcile with the industry’s new AI-driven reality. The coding bootcamps that succeed in the AI era will be those that can prepare students to operate at a more advanced level than is typical for junior engineers. This extends the argument for “role-based” credentialing, which is currently a buzz word in developer certifications. All the major players in the tech certification space (AWS, Salesforce, Microsoft, Google) already lean into this pragmatic and career focused type of skilling credential, and this hyper focus on employability will continue as expectations for developers become more onerous and productivity focused. AI’s disruption to the bootcamp industry runs deeper than merely introducing generalist AI training curriculum; it threatens to upset the skill building industry at its very foundations.
Disclaimer: AWS, Salesforce, Microsoft (GitHub) and Google are all RedMonk clients.
No Comments