Using AI with Exalynt
AI in software engineering sets out Exalynt's stance: AI amplifies software engineering. It does not replace it. This page covers what that means when you work with us: how our engineers use AI on your project, and how you can use AI yourself to make the process easier, whether or not you're technical.
How Exalynt uses AI on your project
Our engineers use AI every day, wherever it helps. On your project, that might look like:
- Getting up to speed: learning your industry's terms, reading through existing code or documents, and drafting the questions worth asking you.
- Showing you ideas early: quick prototypes and mockups you can react to, often within the first days of a block.
- Building: routine code, data migrations, and connections to the tools you already use.
- Checking the work: generating tests and edge cases, and looking for possible bugs and security issues.
- Explaining: documentation, release notes, and checkpoint summaries in your language, not technical terms.
What stays the same
AI changes how the work gets done. It doesn't change how working with us works:
- An engineer is responsible for everything we ship. They understand it, have checked it, and can explain it to you, whether a person or AI produced it.
- Every piece of work still carries a stability level. A quick AI-assisted prototype is labeled Prototype, and nothing is called GA until it has been verified to that level.
- You still buy blocks, not hours. AI doesn't change how you buy engineering. The speed it brings shows up in what each iteration delivers.
- Checkpoints are still about outcomes: what changed, what we learned, and what we recommend next.
Your code and data
- Your policies come first. If your company restricts or prohibits AI tools for its code or data, tell us, and we'll follow your policy. If it changes what a block can deliver, we'll tell you.
- Secrets and real data stay out of AI tools. Passwords, keys, and your customers' real data don't go into prompts, the same way they stay out of code, mocks, and demos at every stability level.
- You can always ask. If you'd like to know which tools we use on your project, or how AI was used in a piece of work, ask, and we'll tell you plainly.
Using AI yourself, with us
We encourage you to use AI throughout a project. You don't need to be technical. AI is good at the tedious parts of getting your thinking clear, and clear thinking makes every block more valuable.
Before you start
- Describe the problem. Talk through your situation with an AI assistant and ask it to help you write a short description: who has the problem, what they're trying to do, what gets in the way today, and what success would look like. Bring that to your first conversation with us.
- Research what already exists. Ask AI which existing products solve problems like yours. If one does, you may not need custom software at all, and that's a good outcome.
- Show what you mean. Use AI to sketch a mockup or a quick prototype of the screens or workflow you have in mind. It's often clearer than a long description.
- List your questions. Ask AI what you should be asking a software engineering team, and bring the questions you care about.
During a project
- Prepare for checkpoints. Jot down what you noticed while trying the latest work, and use AI to turn it into clear, organized feedback.
- Describe real situations. Write examples of how things should work in your business, such as "when a customer cancels within 24 hours, they get a full refund." AI can help you think of cases you'd otherwise forget. These examples are some of the most useful input we get.
- Understand our updates. If a term or recommendation in one of our updates isn't clear, ask AI to explain it, then ask us anything it doesn't settle.
- Plan the next block. Use AI to explore ideas for what to tackle next, and bring the ones you like to the checkpoint.
After delivery
- Understand your software. You own your source code. AI can help you or your team read it and learn how it works.
- Write for your users. Draft help articles, training material, or announcements for the people who'll use what we built.
Bringing us something you built with AI
If you've already built a prototype or tool with AI, bring it. It shows us what you need more clearly than almost anything else. We'll look at it with you and recommend whether to build on it, harden it, or use it as a reference for a fresh start. Like any early work, it starts at Prototype, and we'll tell you what it would take to move it up.
Tips for using AI well
- Give it context. Explain who you are, what your business does, and what you're trying to achieve. The more it knows, the more useful its answers.
- Describe the problem, not just a solution. AI will happily build exactly what you describe. Make sure what you describe is the real need.
- Treat answers as drafts. AI can be confidently wrong, or almost right but not quite. Check facts that matter, especially numbers, legal points, and anything about your customers.
- Keep sensitive information out. Don't paste passwords, keys, or customers' personal data into AI tools, and follow your company's own policies.
- Share the thinking with us. If a conversation with AI shaped your idea, feel free to share it. It helps us understand where you're coming from.
Frequently asked questions
Will you tell me when AI was used on my project?
We don't label every line AI helped with, the same way we don't label which editor we used. But we're open about how we work, and if you ask how AI was used in any piece of work, we'll tell you plainly. We never describe work as reviewed or tested unless it was.
Can you work without AI on my project?
Yes. If your company restricts or prohibits AI tools for its code or data, tell us, and we'll follow your policy. If that changes what a block can deliver, we'll tell you.
Does AI make blocks cheaper?
Block prices are set on exalynt.com, and AI doesn't change how you buy engineering. You buy a block for an iteration, not hours, and the speed AI brings shows up in what each iteration delivers.
Do I need to be technical to use AI with you?
No. Some of the most useful things you can do with AI are describing your problem, researching existing tools, mocking up ideas, and writing examples of how your business works. None of that needs technical knowledge.
Can I bring you a prototype I built with AI?
Yes, please do. It shows us what you need. We'll look at it with you and recommend whether to build on it, harden it, or use it as a reference for a fresh start. See Bringing us something you built with AI.
Who is responsible for code AI helped write?
The Exalynt engineer who ships it, exactly as if they had typed it. They understand it, have checked it to its stability level, and can explain it. That's a description of how we work, not a warranty; your agreement with Exalynt governs any commitments.