Mockmate

Inspiration

Our team brings together students majoring in data science, computer science, and biology.

Preparing for job interviews can be stressful, especially when candidates do not know what questions to expect. While AI chatbots can help people practice interviews, they often ask generic questions and do not fully understand a candidate's background, projects, or the specific position they are applying for.

We wanted to create a more realistic and personalized interview experience. Instead of simply generating interview questions, we wanted an AI interviewer that understands your experience, asks meaningful follow-up questions, and challenges you to explain your work in depth.

That is how Mockmate was born.

What it does

Mockmate is an AI-powered mock interview platform that helps job seekers prepare for real interviews through personalized, interactive voice conversations. Users can customize their interview experience based on their background and target position.

Key features include:

Personalized interview preparation: Users can provide their resumes, job descriptions, GitHub repositories, and project materials. Mockmate uses this information to generate relevant interview questions.
Company-specific interviews: Users can select a company type to practice questions that better reflect different hiring environments and expectations.
Technical and behavioral interviews: Users can choose their interview focus depending on the skills they want to practice.
Real-time voice interaction: Users can speak directly with the AI interviewer, creating a more natural interview experience.
Adaptive follow-up questions: Rather than asking a fixed list of questions, Mockmate can explore a candidate's answers and ask more detailed follow-up questions.
Project-based questions: Mockmate focuses on understanding candidates' actual work and challenges them to explain their technical decisions, contributions, and problem-solving processes.
Interview feedback: After an interview, users can review their performance and identify areas for improvement.

Unlike a general AI chatbot, Mockmate brings interview preparation, background analysis, customized interview settings, and performance feedback into one platform.

How we built it

We started by identifying the limitations of traditional mock interview tools and defining the features that would make Mockmate more useful. We designed a workflow that allows users to provide their background information, customize interview settings, and participate in an AI-powered interview.

Our system uses AI to analyze the information provided by users and prepare relevant interview questions. During the interview, the AI follows the conversation and generates follow-up questions based on the candidate's responses. We also developed a user-friendly interface to make the experience feel closer to a real interview rather than a standard chatbot conversation.

Challenges we ran into

One of our biggest challenges was making AI interviews feel natural and personalized rather than scripted. A good interviewer should not simply move through a list of questions. They should listen carefully, understand the candidate's answers, and ask relevant follow-up questions.

Another challenge was connecting different parts of the experience, including user-provided materials, interview customization, real-time conversations, and feedback. Since we were working within a limited hackathon timeframe, we also had to prioritize our features and balance functionality with usability.

Accomplishments that we're proud of

We are proud of turning our idea into a working mock interview platform during the hackathon. Rather than building another general-purpose AI chatbot, we focused on creating a tool designed specifically for interview preparation.

We are especially proud of our approach to personalized and project-based interviews. Mockmate is designed to help candidates practice explaining what they have actually built, why they made certain decisions, and how they solved problems.

We are also proud of working together as a team with different academic backgrounds and combining our skills to create a product that addresses a real problem many students and job seekers face.

What we learned

Throughout this project, we learned that building an AI application involves much more than connecting an AI model to a user interface. We had to think carefully about user experience, conversation flow, personalization, and how to make AI-generated questions genuinely useful.

We also gained experience collaborating across different technical backgrounds, dividing responsibilities, solving unexpected problems, and making decisions under time pressure. Most importantly, we learned how to turn a common problem into a practical AI-powered solution.

What's next for Mockmate

We hope to continue improving Mockmate by making interviews even more realistic, personalized, and helpful.

Some features we would like to explore include:

Different interviewer personalities: Allow users to practice with friendly, strict, or highly technical interviewers.
More detailed performance analytics: Provide deeper feedback on communication, technical knowledge, and areas for improvement.
Interview progress tracking: Help users monitor their improvement across multiple practice sessions.
Expanded company-specific preparation: Support more detailed interview experiences based on different companies and roles.
Improved interview simulations: Make conversations more natural and closer to real hiring interviews.

Our long-term goal is to make Mockmate a place where job seekers can practice confidently, understand their weaknesses, and walk into real interviews better prepared.

Built With

githubapi
next.js
openaiapi
react
tailwindcss
typescript
vercel

Submitted to

Gator Hack V - AI Days 2026

October 26 at 6:00am EDT to vote

Vote for this project!

Created by

I owned MockMate’s frontend and QA. I built the four-step experience: from entering a résumé and job description, to previewing the interview plan, joining the live interview, and reviewing evidence-based feedback. I connected the interface to our analysis and report APIs, added clear loading and error states, and made the experience work on mobile. I also tested the demo flow and helped catch usability issues before submission.

Xinyu Shen
As Product, Content & Pitch lead, I created Albert, our AI interviewer, and wrote his technical and behavioral personas and the scoring rubrics behind our evidence-based feedback. I researched official Google and Amazon hiring sources for our company-style guides. I also built the demo materials (a fictional company, job description, and resume) and drafted the README, demo script, and Devpost submission.

Lisa Li
I built the backend and the real-time voice interview (OpenAI Realtime over WebRTC), the GitHub repository reader, and the evidence-checked report, and I handled deployment, cost limits and integration. It was my first time building real-time voice, and getting turn-taking right, like not interrupting candidates and handling silence, was harder than I expected, but I learned a lot.

Junhao Li