At Karat, our mission is to make every interview fair, predictive, and enjoyable. In order to accomplish this, we work with you to develop a role-specific, consistent interview structure that allows every candidate to be evaluated based on the same content, format, and scoring criteria.

By maintaining this uniformity in the interview process, we’re able to provide standardized, objective results that make it easy for recruiters and hiring managers to understand a candidate’s potential, dig deeper into their interview performance, and, ultimately, have all the information needed to quickly and confidently make decisions.

In this guide, we’ll provide insight into how we evaluate a candidate’s performance, how you can review interview results, and how you can use our interview results to build hiring confidence.

Karat Interview Basics

Format

Karat Interviews are live, one-on-one sessions conducted by a specially trained and certified Interview Engineer (IVE). The interview takes place via video conference and leverages our collaborative coding environment (IDE) or Whiteboard.

Content

When you open a new role with Karat, the first thing that we do is review the job description and work with you to extract key information around the level, core competencies, and hiring bar. Using this information, we recommend a specific content strategy uniquely suited to your role profile.

We have an extensive content library of industry-validated questions, covering 130+ roles including front-end, back-end, data, and AI roles, but, if you're looking for something unique, we can always create new roles or custom content.

While there are variations from role-to-role, many interviews feature two primary sections:

Discussion & Analysis Questions
Coding Questions

For example, one common format begins with a 10 minute deep-dive technical discussion followed by a 45 minute coding challenge.

The coding challenge portion consists of up to three questions that build on each other while getting progressively more difficult.

Weighting and Minimum Scores

After selecting the content, you have the ability to adjust the weighting and set a minimum score for each of your selected assessment areas.

With weighting, you can tailor your scoring configuration so that each assessment area has the right impact on the candidate’s overall score.

With Minimum Scores, you can require that candidates demonstrate a minimum proficiency in specific assessment areas in order to pass your hiring bar.

Hiring Bar

Once the interview format and content are in place, it’s time to configure your hiring bar. Your hiring bar is the performance threshold that helps you understand if a candidate has the technical skills necessary to be advanced in your interviewing process.

Using Karat's extensive interviewing data and industry benchmarking analysis, we’ll help your set a bar that tailored to your business and the specific role. This bar is both objective and consistent — ensuring that all candidates are held to the same quality standards.

Generating an Outcome

Karat Interview Engineers (IVEs) are certified on each question that they ask. This means that they know how to frame the questions, ways to prompt for explanations, and what to look for in both the approach and answer.

At the end the interview, IVEs use their observations to fill in a standardized rubric and provide any thoughts on the candidate's performance. The IVEs do not provide an overall score or determine an outcome, instead the candidate’s score is algorithmic generated based on the inputs of the rubric, your weighted competencies, and your calibrated hiring bar.

Reviewing Karat Interview Results

Karat Interview Results provide hiring teams with the big takeaways from the interview — including how the candidate performed relative to your bar, how the candidate performed in key assessment areas, and how the candidate performed relative to other candidates.

Let’s take a look at each of the main components that make up the results:

Top section of the Interview Results page

1. The Interview Outcome

This “word score” provides an instant snapshot of the candidate’s performance. It is the main takeaway of the page and is determined by where the candidate’s Overall Score landed relative to your calibrated hiring bar(s).

For each role, hiring teams can define 2-4 Interview Outcomes. If you’re using Karat’s preset categories, the Interview Outcomes will be:

Do Not Pursue (DNP)
Requires Further Review (RFR)
Invite to Next Round (INTR)
Fast Track (FT)

Note: Karat is committed to upholding interview integrity. If an interview is flagged for suspicious behavior, it will be flagged as a fifth category Outcome Withheld - Integrity Risk, and an Interview Integrity Review banner will appear at the top of the results page. Learn more

2. The Overall Score

This “number score” is the total points that the candidate has earned throughout the interview. It is the sum of the individual scores that were received across the various assessment areas. On the graphic, you can see where the candidate’s score landed relative to your bar(s). It’s simple, the higher someone scores, the stronger their performance.

3. Performance Benchmarking

Using the Overall Score, we can compare the candidate’s performance to other candidates who you've interviewed for this same role. On the module, you'll see the candidate's performance percentile, lines indicating your hiring bar(s), and a gray shadow representing the distribution of all candidate scores.

Benchmarking is a great way to understand the relative quality of a candidate — providing insight into the strength of a candidate compared to your sourcing pool.

Middle section of the Interview Results page

4. Reviewing Interview Outcomes

Directly below the Performance Overview section, you’ll see the prompt: Is this candidate’s performance best categorized as [insert outcome]?” By selecting “Yes or No” and completing a brief feedback module, you’re telling us whether or not you agree with the outcome that was awarded. Your input here is invaluable in helping us improve overall alignment. Your feedback is critical for fine-tuning the hiring bars. Please provide as much details as possible. Submitting this review will not impact the candidate’s status or interview outcome.

For more information about reviewing Interview Outcomes, see Submitting an Outcome Review .

5. Interview Summary

Next up, you’ll find the Interview Summary. This AI generated recap provides a high level overview of the interview and the candidate’s performance plus context into how the outcome was determined.

6. Scoring Breakdown

Moving on, you’ll find the Scoring Breakdown section. In this module, we explore how the candidate performed in each of the Interview’s assessment areas. The individual scores in this section are added together to determine the overall score.

Scanning left to right, you’ll see 4 columns:

Areas Assessed: what was covered in the interview segment
Allocated Score: the candidate’s score for that assessment area mapped over the maximum potential score and the minimum required score.
Allocation: the weighting or indication of how much that area contributes to the overall score
Percentile: how the candidate’s performance compared to your other candidates and all candidates across Karat

Expanded Coding Questions Module

7. Coding Score Details

If your role involves "Coding Questions", you’ll be able to dive deeper into the candidate’s performance by clicking the “+” sign at on the far right side of the row. For each coding question that was completed or attempted, we evaluate the candidate for the following criteria:

Completeness: Did the code fully solve the given problem, including edge cases?
Handholding: To what extent did the candidate require hints or guidance from the IVE?
Optimality: Was most efficient solution proposed, considering both time and space complexity?
Time and Space Complexity: Did the candidate correctly identify both, only one, or neither?
Debugging: How much debugging did the candidate have to do on the problem?

Candidate performance in each of these areas directly impacts their “Coding Questions” score. The only exception is Debugging, which is included as a data point but is not factored into the score.

In the example above, we are only showing “Question 1”, but the same details would be available for “Question 2” and “Question 3” (where applicable).

Final section of the Interview Results page

9. Supplementary Insights

The 2nd to last module provides hiring teams with two extra data points that work to provide a bit more context but, ultimately, do not factor into the Interview Outcome. Let’s look at both:

Communication

The candidate’s ability to communicate clearly and effectively as determined by:

Clarity of Speech: How intelligible were the candidate’s spoken words?
Quality of Explanation: How well did the candidate explain their thoughts?
Comprehension: How well did the candidate understand the interviewer?
Interaction: How professional was the candidate’s interaction with the interviewer?

Coding Style

The candidate’s general ability to write code as determined by:

Fluency: How fluent was the candidate was with their language of choice?
Style: How well-structured and readable was the candidate’s code?
Syntax: How familiar was the candidate with the chosen language’s syntax?

Again, neither of these categories factor into the Outcome, but they can serve as a helpful data point for better understanding the candidate’s overall performance and potential.

Exploring Further

Questions and Answers

In the “Details” tab, you have access to each question that was asked by the Interview Engineer, each answer that was given by the candidate, and any notes the IVE may provide.

For the coding exercises, we will also provide an overview of the candidate’s approach, the final version of the coded solution, and any IVE thoughts on their process. With this information, you can dig deeper into the performance when reviewing borderline candidates.

Video and Coding Replay

In the “Video” tab, you have access to the full video recording of each interview along with a synchronized version of the coding environment and time-stamped navigation guide.

While there may be important pieces of information throughout the video, we recommend using the time-stamps to focus on those parts of the interview that are likely to provide the most signal. Specifically, the most value is gained from watching the first 3-5 minutes and the last 2-3 minutes of each section. During these windows, you should be able to see how the candidate interpreted the problem, how they framed their approach, and how they solved it.

Additionally, if there are any areas of concern that are noted in the summary or the detail pages, the video allows you to make your own assessments of the situation.

Help us, help you

No matter the outcome, we highly encourage you to use the “Take Action” button to update the candidate status. Not only is this a great way to stay organized, this status provides us with invaluable data that can be used to improve our process and alignment over time.

Wrapping Up

The best way to get familiar with our scoring and results is to check out a few real interviews in action. Just log into Karat and click on a candidate who has completed an interview to get started!

If you run into any questions along the way, don’t hesitate to reach out to the Karat team at support@karat.com.

Navigating to the Interview Results

You can access your interview results in three ways:

Karat Interview Results page
ATS integration
Karat API

Karat Interview Results page

To review Karat Interview Results page, follow these steps:

Log in to the Karat platform.
Navigate to the Active Funnel or Interviews section.
Click the name of the candidate who has completed their interview. A list of candidates ready for review will appear.
- Optional: Use the Search bar to find a specific candidate by name or email.
You will be redirected to the Karat Interview Results page.

ATS integration

If your organization enabled an ATS integration, you will receive interview results summaries in your ATS integration with enhanced score breakdown and the integrity check incorporated. The integrity check is part of the interview results summary only if an interviewer flags any potential misconduct or cheating during an interview.

Your ATS integration notes will look like the following:

Interview: https://<your-organization>.karat.io/candidates/s/7b5a77fc-5417-4d75-be2f-0be318d81805

Recommendation: <candidate's-name> scored above your standard bar

Total Score: 334/600

Standard Bar: 278

React: 52/130

Develop and Update Backend Code: 282/470

For more information, see ATS integrations overview and How to: Receive status and result updates in your ATS.

Karat API

You can get candidate's interview results through the Karat API. For more information, see the Karat API documentation.

If you have any questions about the changes or need assistance, feel free to reach out to our support team.