How to Talk About AI Experience in Interviews (When You're Still Learning)
You're in an interview. The question comes up: "Tell me about your experience with AI."
You've used ChatGPT. Maybe built a simple project. Taken an online course. But you're not an AI expert. How do you answer without either:
- Overselling yourself (then getting caught)
- Underselling yourself (seeming out of touch)
Here's how to talk about AI in interviews honestly—in a way that actually helps you.
The Mistake Everyone Makes
Bad answer: "I'm very experienced with AI. I use ChatGPT all the time and I'm proficient in prompt engineering and various AI tools."
Why this fails:
- Too vague (everyone uses ChatGPT)
- Claims "proficiency" without proof
- Interviewer will dig deeper and expose gaps
Another bad answer: "I don't really have AI experience. I've just played around with ChatGPT a bit."
Why this fails:
- Undersells yourself
- Sounds uninterested or out of touch
- Misses opportunity to show learning agility
The right approach: Be specific about what you've done, honest about what you haven't, and clear about what you're learning.
The Framework That Works
Answer structure:
- State your actual experience level
- Give specific examples of what you've done
- Show understanding of fundamentals
- Demonstrate learning trajectory
- Connect to the role
Example answer (for someone with basic AI experience):
"I'm early in my AI journey but actively building skills. I've used GPT-4 and Claude to streamline my workflow—specifically, I built a Python script that uses the OpenAI API to analyze customer feedback and categorize issues. That project taught me about prompt engineering, API rate limits, and how to handle inconsistent AI outputs.
I've also taken fast.ai's course to understand ML fundamentals, which helped me grasp when AI is appropriate versus when traditional approaches work better.
For this role, I know I'd be working with [specific AI application from job description]. I don't have direct experience with that yet, but I learn quickly and I'm genuinely excited about AI applications in [industry/domain]."
Why this works:
- Honest about experience level
- Specific project with technical details
- Shows understanding of concepts (not just "I used ChatGPT")
- Demonstrates learning
- Connects to the role
How to Talk About Different Levels of AI Experience
Level 1: "I've Used AI Tools"
Your situation:
- You've used ChatGPT, Claude, or similar
- Maybe for work tasks (writing, research, etc.)
- No programming or technical projects
How to talk about it:
"I've integrated AI into my daily workflow. For example, I use ChatGPT to [specific task] which improved my [metric] by [amount]. I've learned that AI works best when you [specific insight you've gained]. I'm interested in understanding more about AI capabilities and limitations, especially as they relate to [this role]."
Key points:
- Specific use cases
- Measurable improvements
- Lessons learned
- Interest in learning more
What NOT to say:
- "I'm an expert in AI"
- "I know all about prompt engineering"
- Vague claims without examples
Level 2: "I've Built Something with AI"
Your situation:
- You've built a project using AI APIs
- Maybe a chatbot, document analyzer, or similar
- Basic understanding of how AI works
How to talk about it:
"I've built [specific project] using [specific technology]. The project [what it does] and [outcome/result]. The biggest challenges were [technical challenges you faced] and I solved them by [what you did].
This taught me about [technical concepts: API integration, prompt engineering, error handling, etc.]. I also learned about AI limitations—like [specific limitation you encountered] which taught me that [insight].
I'm now learning about [next thing you're studying] to deepen my understanding."
Key points:
- Specific project details
- Technical challenges faced
- What you learned
- Honest about limitations
- Clear learning path
Example:
"I built a document Q&A system for my study group using LangChain and OpenAI's API. It lets students upload course materials and ask questions. I learned a lot about chunking strategies, embedding models, and how retrieval quality impacts answer quality.
The hardest part was handling documents with tables and images—the system didn't parse those well. I solved it by preprocessing documents and adding metadata. The system now answers basic questions well, but I learned it's not a replacement for actually reading the material—more of a study aid.
I'm now learning about fine-tuning and RAG optimization to improve response quality."
Level 3: "I Have ML/AI Technical Background"
Your situation:
- You've studied ML formally (courses, degree)
- Understand algorithms and theory
- Maybe limited production experience
How to talk about it:
"I have a foundation in machine learning from [coursework/degree]. I understand [list key concepts: supervised/unsupervised learning, common algorithms, model evaluation, etc.].
I've implemented [specific algorithms or projects], where I [what you did]. My experience is primarily academic/personal projects rather than production systems, but I understand the challenges of deploying ML at scale from [how you know: coursework, reading, research, etc.].
For production experience, I'm eager to learn about [specific production challenges relevant to the role: model monitoring, serving infrastructure, A/B testing, etc.]."
Key points:
- Clear about your knowledge level
- Specific technical competencies
- Honest about production experience gaps
- Eager to learn production skills
The Questions That Will Expose You
Interviewers ask follow-up questions to test if you actually know what you're talking about:
If you claim you "built a chatbot":
Follow-ups:
- "What technology did you use?"
- "How did you handle errors and edge cases?"
- "What was the biggest challenge?"
- "How did you evaluate if it was working well?"
If you can't answer these specifically, you didn't really build it.
If you say you "know prompt engineering":
Follow-ups:
- "What makes a good prompt?"
- "How do you handle inconsistent outputs?"
- "What's an example of a complex prompt you wrote?"
If you can't give specifics, you don't really know it.
If you mention "machine learning experience":
Follow-ups:
- "Explain the bias-variance tradeoff"
- "When would you use regularization?"
- "How do you handle imbalanced datasets?"
If you can't answer fundamentals, you don't have real ML knowledge.
The Red Flag Phrases to Avoid
Things that sound impressive but mean nothing:
❌ "I'm a prompt engineering expert" ❌ "I have extensive AI experience" ❌ "I'm proficient in all major AI tools" ❌ "I've mastered ChatGPT" ❌ "I specialize in leveraging AI solutions"
Why these fail: Too vague. No proof. Everyone can say this.
Better alternatives:
✅ "I've built 3 projects using AI APIs" ✅ "I understand how RAG systems work and have implemented one" ✅ "I use AI tools daily in my workflow and have measured the productivity impact" ✅ "I've completed coursework in ML and can explain fundamental concepts" ✅ "I'm early in my AI journey but actively building skills"
How to Handle "We're Looking for AI Experience" When You Don't Have Much
Situation: The job asks for "AI experience" but you're light on it.
Bad approach: Pretend you have more experience than you do.
Good approach: Acknowledge the gap, emphasize learning ability, show genuine interest.
Example:
"I know this role requires AI experience and I'm honest that I'm still building my skills in this area. However, I'm a fast learner—when I needed to [past example of learning something technical quickly], I [what you did] in [timeframe].
I've already started preparing for this role by [specific things you've done: courses taken, projects built, reading/research]. I'm genuinely excited about AI applications in [domain] and I'm committed to quickly getting up to speed.
What I bring is [your other strong skills that are relevant], and I'm confident I can combine that with AI technical skills as I develop them."
This works because:
- You're honest (builds trust)
- You show initiative (already learning)
- You demonstrate learning ability (past examples)
- You emphasize complementary strengths
The "Show Don't Tell" Strategy
Instead of claiming skills, demonstrate them:
Weak: "I'm good at using AI tools."
Strong: "I automated our team's meeting note-taking using ChatGPT's API, which saves us about 5 hours per week. I set it up to summarize recordings, extract action items, and categorize decisions. It required handling API authentication, managing costs, and refining prompts through iteration."
Weak: "I understand machine learning."
Strong: "For my capstone project, I built a churn prediction model using random forests. I handled class imbalance with SMOTE, used cross-validation for evaluation, and achieved 84% precision and 79% recall. The toughest part was feature engineering—I tried 15+ features before finding the combination that worked."
The principle: Specifics beat generalities every time.
Questions You Should Ask Them
Flip the script and ask about their AI:
Good questions:
- "How is your team currently using AI?"
- "What AI challenges are you facing?"
- "Where do you see AI going in your product roadmap?"
- "What AI skills are most important for success in this role?"
- "How do you evaluate if an AI feature is ready to ship?"
Why this works:
- Shows you're thinking strategically
- Reveals what they actually need
- Demonstrates genuine interest
- Gives you information to assess the role
The Honesty Test
Before the interview, ask yourself:
Can I explain this in technical detail for 5 minutes?
- If yes → you can claim this experience
- If no → be more modest in your claims
Could I do this again from scratch?
- If yes → you know this
- If no → you followed a tutorial (be honest about that)
Can I discuss what went wrong and how I fixed it?
- If yes → you have real experience
- If no → you might be overstating
The rule: Only claim experience you can back up with specific details.
The Growth Mindset Answer
For roles where you're underqualified:
"I know I don't have [specific experience they want] yet. But here's what I've learned about myself: [example of quickly learning something difficult].
I'm the type of person who [your learning style: reads documentation thoroughly, builds projects to learn, asks questions, etc.]. For this role specifically, I've already [concrete prep you've done].
What excites me about this position is [genuine reason]. I'm not looking for a role where I already know everything—I want to be challenged and grow. And I'm confident I can get up to speed quickly because [evidence from past experience]."
This works when:
- You're genuinely willing to learn
- The role has training/mentorship
- Your other skills are strong
- You can prove you learn fast
The Bottom Line
When talking about AI in interviews:
DO:
- Be specific about what you've actually done
- Share technical details and challenges
- Be honest about your experience level
- Show what you're actively learning
- Connect your experience to the role
DON'T:
- Claim expertise you don't have
- Use vague buzzwords without specifics
- Pretend you know things you don't
- Undersell legitimate experience
- Ignore AI entirely (it's too important)
The goal: Come across as someone who:
- Has genuine interest in AI
- Is actively learning
- Can talk intelligently about what you know
- Is honest about gaps
- Will continue growing
That's more valuable than pretending to be an expert when you're not.
---
Preparing for interviews and not sure how to position your AI experience? JobsLoop helps you understand what companies actually expect—so you can prepare honest, compelling answers.