You submit your STAR response broken into four fields—Situation, Task, Action, and Result—along with metadata (company, role, level, and URL). The model uses that structured input to drive its evaluation.

We feed your STAR text into a seven-step, chain-of-thought prompt that checks key gates (sentence counts, illustrations and data, and job context), reasons through 13 distinct scoring elements, and generates targeted feedback for each of your STAR sections.

You’ll get 6-8 sentences of narrative feedback for each of the four STAR sections—plus an overall summary listing your top strengths, weaknesses, and 3–4 concrete “next steps” to improve.

Yes. In fact, we encourage it. You can take the feedback from your first submission, revise your response accordingly, and submit a new version to see how your score and feedback improve. Each submission is scored independently, and the history table tracks your progression over time.

Through our API backend, OpenAI call, and deterministic scoring pipeline, results—including email notifications—are delivered within seconds of your submission.

o 1–49: Needs improvement—your story is missing key structural elements or data.
o 50–79: Acceptable—you’ve covered the fundamentals but still have flaws you should address.
o 80–100: Great—your response is “shippable,” demonstrating strong structure, data, and impact that would impress most interviewers.

Two main factors can introduce slight variability:
o AI nondeterminism: Even with the same prompt, the model may emphasize different word choices or examples each call.
o AI Calibration: Your raw score is compared to human-scored anchors and evaluated within ranges of these anchors, which can shift your final number up or down by several points (but not materially enough to push you in a different score range).

A score right around 50 means you’ve hit the minimum bar but still have clear weaknesses. Usually using data points or tightening your illustrations would move you solidly into the “acceptable” range. You should follow the customized feedback from our model to help you get there.

Usually focusing on data depth and clear behavioral illustrations will push your score to 80+. Follow the specific feedback from our model to finetune your STAR response into a shippable state.

Yes. Your STAR response is evaluated in context with the job title, level, and company you’re targeting. This allows the model to tailor its scoring and feedback based on expectations for the position—especially if you’re applying to companies known for high-bar behavioral interviews like Amazon, Microsoft, or Google.

Once you’ve used all your STAR submissions for the month, the “Submit” button will be disabled, and you’ll receive a prompt encouraging you to upgrade or wait for your next billing cycle. You’ll still be able to view your past scores and feedback anytime on your dashboard.

We take data security and privacy very seriously. All submissions travel over HTTPS with TLS encryption. Once in our system, your data is encrypted at rest by default. We adhere to GDPR and CCPA guidelines. You can view our full Privacy Policy here.

We offer one-on-one coaching packages that build on your AI-driven feedback with live, human-led guidance. To get started review our coaching packages here or email us at service@koalanda.com.