A Look at General Assembly’s New AI Software Engineering Pathway

As AI reshapes the skills required in tech roles, General Assembly is evolving how developers learn and upskill. The new AI Software Engineering Pathway replaces the traditional one-size-fits-all bootcamp with a flexible sequence of stackable courses designed to build foundational coding knowledge and apply it to modern AI-powered applications. Madi Coates, Manager of Content at General Assembly, and Beatrice Partain, Director of Product Management at General Assembly, explain what’s driving this shift, break down the four courses in the pathway, and share how GA is adapting its programs to support continuous learning as roles across tech continue to evolve.

General Assembly recently introduced the AI Software Engineering pathway as part of a broader shift toward AI-focused learning – what’s driving this evolution?

Every role is changing. For the last 10 years, even though General Assembly taught our software development bootcamp that served so many students well, we never saw a need for mid-career skills for software developers, because most software engineers trained on the job. We always focused on the entry-level software developer.

AI is the rapidly changing technology that is disrupting the way people work and exist in the world. We are seeing more job opportunities that are expecting some level of competency in using AI. GA has always followed the skills needed to stay competitive in the job market. As the job market and required skills have changed, we are also changing our approach. Our new courses are not 420 hours or even 40 hours – they are 32 hours, running across multiple time-zones so learners can find a time that works for them. This is because AI allows you to do more in less time. It's an opportunity to adapt to what technology requires in the workplace and to use that technology to learn more in less time.

What do these upskilling shifts look like in the real world? 

This is actually something I'm very passionate about. Many people at GA are career transitioners. My shift was small – project management to product management – but the introduction of AI is changing how roles expand and morph. A UX designer is now asked to build HTML prototypes. Product managers are asked to prototype. Project managers are asked to make data visualizations. Software developers are asked to know a little about UX design. Digital marketers are asked to know a little about UX. Roles are no longer rigid; they are more like a living, breathing organism that's changing. AI is both influencing and allowing this to happen. It's the forcing mechanism and the tool of this change.

It's no longer safe to say that everyone who is a product manager, data analyst, or software developer needs the same particular skills. It becomes much more about you, your role, and your company's blending of roles. What skills do you need to move onwards in your career? That's how we're thinking about it. We have clearly outlined pathways — what we think you need if you want to go in a certain direction — and you as the learner can also customize that experience to make your individual upskilling plan.

How does General Assembly help learners customize the pathway based on their career goals or existing skills?

Every individual who speaks to a GA admissions person will get individualized support: "What courses make sense for you in your career?" It really is hyper-individualized at this point. It's no longer safe to say that 45 people all need the same 400 hours of training.

We want to meet everyone where they're at. We want everyone who shows up at GA to be able to get the skills they need without also getting a bunch of skills they don't need. It's hyper-targeting.

What are the 4 courses that lead into the AI Software Engineering pathway?

The first two courses are fundamental: Frontend Development with HTML and CSS and Backend Development with JavaScript. AI software engineering still requires a foundational understanding of code – how to build, review, edit, and debug. AI is a tool, not a magic wand, and a strong foundation is essential for its appropriate and ethical use.

Frontend Development with HTML and CSS provides the structural and visual foundation of the web, focusing on layout and responsive design. Backend Development with JavaScript introduces the 'brain' of the application, covering server-side logic, APIs, and databases to power the user experience. These lead to AI Web Applications, which marries the two – front-end and back-end – to design web applications, acknowledging that most modern products will include an AI component.

The final course, AI Systems Engineering and Reliability, is shared with the AI and Machine Learning pathway. It addresses the growing need for software engineers to understand data, as AI and machine learning become increasingly intertwined. Reliability and security become paramount when working with more data and large language models.

Who is this pathway designed for – career switchers, working developers, or professionals looking to layer AI skills onto existing roles?

We've been joking that the persona is us – you, me – and it is true. We're talking about software developers, but front end development with HTML and CSS is relevant for UX designers and product managers too.

The first two courses are ideal for junior individuals considering software development. The third is great for entry-level people focusing on web development. Finally, the AI Systems Engineering and Reliability course is for existing software developers whose careers are being directly impacted by the increasing role of AI in their job.

We're targeting a broad base of people – anyone who needs a specific skill rather than someone looking for a job in a specific field. Take front-end development. It's a junior skill for a software developer, but not for a UX designer. What is entry-level? What is mid-career? 

GA students are often new to their role, but not new to work. If you worked for 10 years in another industry and shift, you're entry level in that specific skill/role, but not new to the workforce. You're an established professional with transferable skills. That's what we're leaning into. AI, and the blend of roles, will make this even more true.

The beginning, middle, and end of a career won't look the same, and I think that's good. Roles that people will be in in two years don't exist yet. It becomes a matter of what the next skills are, rather than how to forever proof your career.

Many career-switchers worry about learning AI without a strong technical background – how does the pathway build confidence and foundational software engineering skills before layering in AI?

First, fundamentals are always important. You need to understand the technology in order to use it responsibly. We start the pathway with the foundations of coding without the use of AI. As the learner progresses, we begin to introduce using AI as an assistance. They learn how to discern the quality and accuracy of the output. Then they begin to work collaboratively with the tool to support them in debugging code to learn from their mistakes without losing momentum. They learn to iteratively make improvements with each interaction with the tool. Through the pathway they are becoming confident in judging, validating, and refining AI-generated code. Like I said earlier: AI is a powerful tool, but it’s not a magic wand – humans still have to review, question, and take responsibility for the output. I like to think of AI tools as interns: they can help you move faster, but you still have to sanity-check their work.

At the same time, the more AI speeds up technical work, the more important human skills become. What people often call “soft skills” – communication, judgment, adaptability – are really the things that make you uniquely valuable. Our goal is to help learners develop the discernment to know when to rely on AI and when to rely on their own expertise. That flexibility will be essential as technology continues to evolve.

Ultimately, we want students to feel confident collaborating with AI rather than fearing it. AI can do some things extremely well, but there are also many things humans do far better. Learning how to work alongside these tools – while still bringing your own critical thinking and perspective – is what will help professionals stay adaptable in the years ahead.

How do the projects in the AI Software Engineering pathway reflect how modern engineering teams actually build and deploy AI-enabled products?

These projects are even more hands-on.

All of our new programs include two key elements: a capstone project and 50% lab time. You will leave with a tangible project you've built, and half of your class time will be dedicated to hands-on work. For a two-hour class, expect one hour of curriculum and instruction, and one hour of practical, hands-on time at your keyboard. This emphasis on doing is intentional because we understand that it's challenging for professionals – our clients, potential students, and expert instructors – to take time out of their day to learn. It's vital that learning happens as part of the process of doing.

Our historical bootcamp learning model is the "I do, we do, you do" approach. An instructor explains a concept (I do), you practice it together in a lab (we do), and then you apply it to your capstone project independently (you do). This model ensures highly practical, application-focused learning, which aligns with sound learning science, especially for adult learners. This portfolio takes that approach and applies it to shorter courses, but the foundations of the learning experience remain the same.

Our learning experience is not randomized; it is intentionally designed based on adult learning philosophies and current research. The structure and look of our courses are research-backed to ensure effective learning.

The final course covers AI systems engineering and reliability – why is it critical for developers to understand deployment, monitoring, and performance, not just model integration?

Because AI integrates everything, our work processes are increasingly tied together. It's no longer enough to look only at your piece. The way data must flow back and forth for AI models to function makes isolation very difficult. You are only as strong as your weakest link. Security gaps in your software engineering workflows will compromise your software, regardless of the strength of other pieces. If that makes sense. now interconnected, but will have a better answer for you.

You need the skills and discernment to know what you're talking to, what it knows, and what it doesn't. Models are getting better at working with you, but having discernment around the answers you receive at every process point is still important.

For learners who complete the full pathway, what kinds of roles or responsibilities are they better prepared to step into compared to traditional software engineering graduates?

Every role is changing, right? Salespeople are being asked to be more technical; technical people are being asked to sell; product managers are being asked to master data visualization. It is becoming increasingly difficult to nail down exactly which fixed roles will exist in five years.

But there is freedom in that uncertainty. Traditionally, the smoothest path for a career changer was to transition within their own organization—a product manager becoming a developer in the same department, or a legal professional becoming a data analyst for a law firm. The logic was simple: the fewer steps you took away from your domain expertise, the easier the transition.

Today, this new flexibility—driven by AI—allows learners to expand their skills without waiting for a formal title change. For instance, a product manager can now 'inch' toward software development by gaining technical experience directly within their current role.

This pathway prepares learners for this new reality. Beyond just coding, it equips them with the ability to act as a founder. When you have a toolkit that supports you across the board, you don't have to wait for someone to 'give you a shot.' You can build the app yourself. You can leverage LLMs to draft product requirements or use AI agents to handle visual design and prototyping. By mastering 'vibe coding' and cross-functional AI skills—the core of our AI Fundamentals pathway—our graduates aren't just looking for a job description; they are ready to build the roles they want to inhabit.

How does this lower the barrier to entry for learners who may be curious about AI but not ready to commit to a full program?

This portfolio's mission is all about "choose your own adventure." That's not meant to be scary; it's freeing. It allows you to take charge of your career direction and inspires opportunity rather than restriction.

It offers so much pivoting opportunity. Say you start with front-end web development, convinced you want to be a software developer. If you find front-end, perhaps with CSS, more engaging than back-end, you can pivot to design next. You aren't committed forever to where you started. There's ample opportunity to pivot and iterate on your career and learning goals as you go. That feels exciting to me, especially since these courses are designed to fit into a mid-career life. You don't need to pause your life for 12 weeks, as you would with an intense bootcamp. These are meant to fit around your life, with your current job. You can try something new without stopping what you're doing, even if you just start with one course instead of four. And for those that need or want an even smaller bite size experience before enrolling in one of our pathway courses, we have workshops that also feed into these pathways, like our Front-End Basics workshop.

The way we help our students reach their next goal hasn't changed. Only the goals and the path to them are different.

Looking ahead, how does this pathway fit into General Assembly’s broader vision for AI education across disciplines such as product, design, and business?

We’ve always got our eyes on the horizon, towards what skills and training people will need next, and that includes software developers. I think it’s naive to know what skills we’ll be training on a year from now, because some of those skills don’t even exist yet, so I think it’s safe to say that our vision for AI education isn’t all that different from what we’ve always done - helping our students pursue skills and work they love. What that looks like may change, but that’s always the foundation.

Find out more and read General Assembly reviews on Course Report. This interview was produced by the Course Report team in partnership with General Assembly.