AI, Natural Language Processing and the Future of Computer Science Careers
Large language models are changing the way software gets built. Strong computer science foundations matter more as a result.
Dr. Curry Guinn
Teaching Professor, CU Boulder Applied Computer Science Post-Baccalaureate Program
Natural language processing, or NLP, is the area of computer science focused on helping computers work with human language. That includes reading, classifying, translating, summarizing, generating, and responding to text.
NLP is suddenly everywhere
For a long time, natural language processing was a specialized area of artificial intelligence. It helped computers work with human language: classifying text, translating sentences, searching documents, summarizing information and finding useful patterns in words.
That work is now part of everyday computing. You see it when you use a chatbot, ask a search engine a question, summarize a document, translate a message, dictate into your phone or use an AI coding assistant.
Natural language processing has moved from a specialized research area into the tools people use every day.
Large language models have made language feel like a new interface for computing. You can ask for code, request an explanation, compare documents, draft a plan or get help understanding a technical error.
That is a real change. It also raises a reasonable question for students: if AI systems can write code, answer technical questions and explain complex ideas, does computer science still matter?
Computer science still matters
The answer is yes. The reason it matters is changing, and that is where the public conversation often gets too simple.
One version says AI will replace programmers. Entry-level jobs will disappear. A computer science degree will no longer be worth it.
That version is too simple. The current job market is real, and students are right to pay attention to it. Hiring has changed, employer expectations have changed, and students entering the field need to think carefully about the skills they build and the projects they can show.
Recent workforce data point in the same direction. CompTIA’s Tech Jobs Report, drawing on Lightcast job-posting data, showed tech job postings rising through the opening months of 2026, with software developers and engineers among the roles in highest demand. That evidence does not support the simple story that software work is disappearing. A better reading is that the skill mix is changing.
Employers still care whether applicants can write code. They are also asking whether applicants can work with AI tools, data systems, machine learning workflows and language technologies. Most students will use AI as part of broader technical work rather than becoming AI researchers. That makes AI literacy a useful part of the toolkit.
Computer science is not disappearing. The work is changing.
One thing to keep in mind is that the shift seems to be happening at the level of skills. Job titles may look familiar: software engineer, developer, data analyst, systems engineer. The expectations inside those roles are changing.
NLP is one area within artificial intelligence. AI is the broader field of building systems that can perform tasks we usually associate with human intelligence, including computer vision, robotics, self-driving cars, recommendation systems, computer-generated art and music, speech recognition and natural language processing.
NLP sits inside the larger field of artificial intelligence, with strong connections to machine learning and many other AI areas.
When code is easier to generate
AI coding tools make it easier to produce code. That matters because a student, analyst, researcher, or small business owner can now create small programs that would have required much more technical knowledge a few years ago.
Generating code is one step. Building reliable software is a larger problem.
Someone still has to decide what the system should do. Someone has to understand the data, test the output, find the edge cases, think about security and decide whether the generated code fits the larger system.
At that point, judgment starts to matter more than typing syntax.
When code becomes easier to generate, judgment becomes more important.
That is where computer science foundations matter. Algorithms and data structures help you reason about efficiency. Databases help you understand how information should be stored and retrieved. Software engineering helps you build systems that can be maintained and changed over time. Computer systems help you understand what is actually happening underneath the code.
AI, NLP, and machine learning build on a broader foundation. Students still need the core ideas behind programming, algorithms, data, systems, software engineering, and mathematical reasoning.
AI tools can help with many tasks. They still leave students with the responsibility to understand what they are building.
Why NLP and LLMs matter for students
Natural language processing is especially important because it is near the center of the current AI shift. Large language models are language systems. They are also coding assistants, writing assistants, search tools, tutoring tools, analysis tools and workplace automation tools.
Most students will not become machine learning specialists. They still need to understand enough about AI, machine learning, data, and NLP to use these systems well.
In practice, that means asking better questions. What data was the system trained on? What does the model actually know? How should the output be evaluated? What happens when the model is wrong? How does this tool fit into a larger workflow?
Those are technical questions, and they are also professional questions. The strongest students will be the ones who can use AI tools while still thinking carefully about correctness, reliability and context.
Earlier NLP systems often handled narrow tasks. Large language models made language interaction more flexible and conversational. They can write, explain, summarize, translate, generate code and respond across many different contexts.
Teaching NLP in the Computer Science Post-Bacc Program
I’ve been working in natural language processing for more than 35 years, beginning with my PhD work at Duke University and continuing through applied NLP work in industry at RTI International. During that time, I’ve seen the field move through several major phases: rule-based approaches, statistical methods, machine learning, neural networks and now large language models.
That history is one reason I enjoy teaching NLP in the Computer Science Post-Baccalaureate (CSPB) program. Students are stepping into a long-running area of computer science that now sits near the center of how people use AI.
NLP did not begin with ChatGPT. What changed is how visible, accessible, and powerful these systems have become.
In the NLP course, students build projects that show how broad the field has become. Some projects focus on classification, summarization or question answering. Others apply NLP to finance, law, medicine, cybersecurity, creative writing, software documentation and public communication.
The best projects usually combine technical skill with a real problem the student understands. They also require habits that matter in professional work: defining the task clearly, working carefully with data, choosing and comparing methods, evaluating results and explaining what the system can and cannot do.
Why CSPB fits this moment
Students in CSPB’s NLP course have applied language technologies to public-facing and practical problems, including misinformation and political bias detection, toxic comment filtering, financial sentiment analysis, podcast and news summarization, question-answering tools, legal document simplification, biomedical named entity recognition, cyber threat intelligence extraction and text-to-SQL generation.
The strongest projects do more than use an AI tool. Students define a language problem, work with real text data, compare methods, evaluate results, and explain the system’s limits.
The CU Boulder Applied Computer Science Post-Baccalaureate program is built around an interesting kind of student. Every student in the program already has a degree in another field before studying computer science.
That matters.
In software work shaped by AI, domain knowledge becomes more valuable. Healthcare software needs people who understand healthcare. Financial systems need people who understand finance. Educational technology needs people who understand teaching and learning. Scientific computing needs people who understand the science.
CSPB students often bring something valuable into computer science: another field of knowledge.
The program adds the computer science foundation to that earlier experience. Students learn programming, algorithms, data structures, software development, systems and related areas. They also bring their previous academic and professional backgrounds with them.
That combination fits the moment. AI can generate code, and people still need to frame problems, evaluate outputs, understand users and build systems that work.
- Programming
- Algorithms and data structures
- Discrete math
- Probability and statistics
- Computer systems
- Databases
- Programming languages
- Software engineering
- Data science
- Machine learning and NLP
What students should build
For students thinking about computer science now, my suggestion is to build in two directions at once.
First, build the foundation. Learn to program. Learn data structures and algorithms. Learn databases, systems, programming languages and software engineering. Those ideas are still the core of reliable software work.
Second, add the modern layers. Spend time with AI tools. Learn the basics of machine learning and natural language processing. Build projects that use AI carefully. Be able to explain what the system does, how you evaluated it and where its limits are.
Projects matter here. They give you something concrete to show. They also help you practice the kind of judgment that AI tools cannot replace.
The future is still being built
AI and NLP are changing computer science, and students should take that seriously.
The change makes preparation look different.
The future will need people who can work with AI systems, evaluate their outputs and understand the systems around them. It will need people who can connect technical foundations with real problems in real domains.
That is a strong reason to study computer science now, especially for students who want to understand AI rather than simply use it.
Students in the CSPB program already have a degree in another field. In the AI era, that prior background can become part of their technical strength. The goal is to connect computing with real problems.