Software Engineering

What Do I Need To Know?

Software engineering is certainly more than just programming—it’s about the systematic application of engineering approaches to the development of software systems. There are various principles, methods, and best practices that guide the process. Your goal as a software engineer is to create systems that are efficient, maintainable, scalable, reliable, and secure.

To build a software system, you’ll need to know exactly what the user requirements are, so you’ll need to know how to get them, and be able to communicate with stakeholders effectively before, during, and after development. Communication is a power skill.

Here’s what you need to know:

One of the main takeaways from this video: Programming is turning specifications into working software. AI is good at that. Human “programmer” jobs have been going away for a while now—AI coding agents are doing almost all of that work now. Engineering, on the other hand, involves design, reliability, tradeoff analysis, cross functional collaboration, and incident response. Humans are needed for these tasks. The number of software engineering jobs is expected to rise over the next several years.

The SDLC

There are quite a few phases one goes through in building software systems. Roughly, they are Ideation, Planning, (Requirements) Analysis, Design, Implementation, Code Review, Testing, Documentation, Integration, Deployment, Maintenance, Evaluation, Retirement, Disposal. Collectively, these phases are known as the Software Development Life Cycle, or the Systems Development Life Cycle (SDLC). You can learn more by starting a web search at Wikipedia.

Skills and Practices

In no particular order, software developers, engineers, architects, and anyone involved in producing software systems should hone technical skills in:

• Programming languages (you should be able to read and write code in multiple languages)
• Frameworks, libraries, APIs, and tools
• Working with agents
• Version control systems
• Testing methodologies
• Debugging techniques
• Design patterns
• Project management tools
• Working with command line interfaces ALWAYS IMPORTANT

and, of equal or greater importance, so-called durable human skills, also known as power skills (or sometimes soft skills), such as:

• Communication (oral and written, also understanding your audience)
• Collaboration
• Inclusion (bring everyone together, software is a team effort)
• Problem solving (includes first identifying the real problem)
• Critical thinking (this term is overloaded, but hopefully you get the idea)
• Creativity
• Adaptability and flexibility
• Time management
• Organization
• Attention to detail
• Ethical awareness
• Empathy
• Emotional awareness
• Self-management
• Continuous learning

People skills are power skills.

Initiation

The early phases of the SDLC are focused in ideation, gathering requirements, understanding the problem domain, defining the scope of the project, and high-level design.

• Ideation: Identifying an opportunity for an audience. Understanding users and their needs.
• Conceptualization: Exploring potential solutions. Coming up with multiple approaches and evaluating their risks and feasibility. Identifying the systems boundary. Performing cost-benefit analyses. Generating a proposal.
• Planning: Outlining the project’s goals, timeline, resources, and budget. Defining who the stakeholders are and their roles.
• Requirements Analysis: Gathering and analyzing user requirements to ensure the software meets their needs. For large projects, you need more that a requirements document, you need a requirements database. Includes not only what the project needs to be correct (functional requirements), but also what it needs to satisfy cost, latency, and privacy concerns.
• Design: Creating the architecture and design of the software system. Where will the code run? What kind of storage is necessary? What kind of network infrastructure is required? What kind of security measures are needed? (You’ll need to learn about Threat Modeling.) How will users interface with the system? What are the correctness criteria?

Notice that security concerns are part of system design, and not something you tack on at the end.

Security is a design concern, not an afterthought.

Implementation

The software that gets built includes (at least) executable programs, supporting scripts, and configuration files. Test scripts are essential during implementation, too! They should be created alongside the main code to ensure functionality and reliability. Code reviews for correctness, conformance, security compliance, and performance are also crucial.

This is the part that AI agents can do pretty well. But humans must be kept in the loop.

As of 2026, four “layers” are emerging in getting the code and tests produced by AI agents:

• Prompt Engineering (a.k.a. Prompt Construction)
• Context Engineering
• Intent Engineering (a.k.a. Intent Alignment)
• Agentic Engineering (a.k.a. Workflow Architecture)

You can find details at: MindStudio and Elastic Search Labs. Or this article with tips on doing good context engineering. For now, here’s a little table of the main ideas:

There are many resources available that will show you, for each layer, best practices (and common mistakes). Mistakes can be quite costly!

Implementation is not just “coding” the main functionality. Testing and Security (especially Software Security) are concerns that are 100% integral to the development process.

Testing happens during software development.

Ethics

The two major professional societies for software engineers are the IEEE Computer Society and the ACM, and each have published codes of ethics. They should be read in full:

• ACM Code of Ethics and Professional Conduct
• ACM Software Engineering Code of Ethics and Professional Practice
• IEEE Computer Society Code of Ethics

If you like bullet point summaries, we got you covered! Read on.

Key principles (an oversimplification, but a good place to start):

• Act in the public interest
• Maintain professional competence
• Protect confidential data
• Avoid algorithmic bias
• Foster fair, collaborative working environments

Core Ethical Principles:

• [Public] Act consistently in the public interest and welfare.
• [Client and Employer] Act in the best interests of clients and employers, while maintaining ethical standards.
• [Product] Build high-quality products that meet professional standards.
• [Judgment] Maintain integrity and independence in professional judgment.
• [Management] Promote ethical approaches when managing and leading software teams.
• [Profession] Advance the integrity and reputation of the software engineering profession.
• [Colleagues] Be fair and supportive of colleagues.
• [Self] Participate in lifelong learning and ethical practice.

Themes:

• [Bias Mitigation] Actively mitigate prejudices within datasets to avoid discriminatory software, particularly in recruiting, finance, and health care.
• [Privacy & Data Protection] Safeguard user data and adhere to privacy laws.
• [Safety & Security] Develop robust, secure systems to prevent exploitation, confidentiality breaches, data update or destruction, and system failures.
• [Ethical Design] Create systems that respect user boundaries and enhance well-being.

Professional Responsibilities:

• [Competence] Only take on work within your expertise.
• [Growth] Continuously update your skills and knowledge.
• [Accountability] Take responsibility for your actions and decisions.
• [Transparency] Disclose potential project failures, safety risks, or ethical conflicts.
• [Attribution] Respect copyrights, licenses, patents, and the intellectual work of others.

You can find another short summary at Fellow.ai.

Recall Practice

Here are some questions useful for your spaced repetition learning. Many of the answers are not found on this page. Some will have popped up in lecture. Others will require you to do your own research.

1. What is software engineering?
   The systematic application of engineering approaches to the development of software.
2. What is programming?
   The conversion of specifications into executable code.
3. Marina Wyss mentions that AI is really good at programming, but there are things that humans excel at in the software engineering process. What are those things?
   Design, reliability, tradeoff analysis, cross functional collaboration, and incident response.
4. What are some of the phases of the Software Development Life Cycle (SDLC)?
   Ideation, Planning, (Requirements) Analysis, Design, Implementation, Code Review, Testing, Documentation, Integration, Deployment, Maintenance, Evaluation, Retirement, Disposal.
5. Name several technical skills important for software engineers.
   Languages, frameworks, agent orchestration, version control, testing, debugging, design patterns, command line interfaces, project management.
6. The term “soft skills” has been around a while, but there are at least two better terms. What are they?
   (1) Durable Human Skills and (2) Power Skills.
7. Name as many durable human skills as you can.
   Communication, collaboration, inclusion, problem-solving, creativity, adaptability, empathy, ethical awareness, critical thinking, leadership, time management, emotional awareness, self-management, continuous learning.
8. Is there a security analysis phase of the SDLC? Why or why not?
   Security analysis should be integrated throughout the SDLC rather than being a separate phase. Security must be built-in to the entire software development process.
9. What are the four layers of agentic development?
   Prompt engineering, context engineering, intent engineering, agent orchestration.
10. What two professional organizations provide codes of ethics for software engineers?
    The Association for Computing Machinery (ACM) and the Institute of Electrical and Electronics Engineers (IEEE).
11. What are the 8 principles of software engineering ethics (from the ACM/IEEE code)?
    Public, Client and Employer, Product, Judgment, Management, Profession, Colleagues, Self.

Summary