Inspiration

Our inspiration for Ace.AI comes from our journey as software engineering students grinding LeetCode to land interviews, only to realize that technical skill alone isn’t enough. We saw peers struggle not with solving problems, but with communicating them. That insight sparked Ace.AI: a platform designed to help you master both technical thinking and the way you express it, so you can truly ace your interviews.

What it does

Ace.AI delivers real-time, voice-based interview practice powered by AI. Choose between behavioral or technical modes, experience adaptive questioning that mirrors real interviews, and receive instant, structured feedback on your performance.

How we built it

We built Ace.AI as a full-stack, AI-powered interview platform focused on real-time voice interaction and intelligent feedback. On the frontend, we used React, Typescript and Tailwind CSS to create a clean, responsive interface, integrating the Vapi API for seamless voice experiences.

On the backend, we leveraged Node.js, TypeScript and Express to handle interview sessions, process transcripts, and manage API communication. We integrated Vapi for real-time voice and text-to-speech functionality, enabling natural, conversational interviews.
Once a user completes a response, the transcript is sent to the backend, where the OpenAI API evaluates performance across key areas like clarity, technical accuracy, and communication. Structured feedback is then generated and returned instantly to the user.
Our MVP focuses on delivering a smooth voice-based interview flow with 5-question sessions, basic scoring, and an intuitive UI—laying the foundation for future features like advanced analytics, personalized interview tracks, and live coding simulations.

Challenges we ran into

Merge Conflicts & Team Collaboration Working collaboratively through GitHub led to merge conflicts that disrupted development. We had to quickly learn better version control practices to avoid breaking the project.
Voice API Integration & Configuration Integrating Vapi for real-time voice interaction introduced challenges around timing, delays, and syncing responses. Ensuring the conversation felt natural required multiple iterations and debugging.

Accomplishments that we're proud of

Leveraged Team Strengths Effectively We identified each team member’s strengths early and divided responsibilities accordingly, allowing us to build a cohesive and efficient system under tight time constraints.
Successful Frontend–Backend Integration We connected a React-based frontend with a Node.js/TypeScript backend, enabling real-time communication and a seamless user experience.
Engineered an Adaptive Interview System By designing and refining a series of structured prompts, we developed an interviewer that dynamically adjusts follow-up questions based on user responses. Built a Functional End-to-End MVP Within a short timeframe, we delivered a working product that includes voice interaction, response evaluation, and structured feedback.

What we learned

API Key Management & Secure Integration We learned how to properly work with API keys for services like AI and voice processing. This included securely storing keys using environment variables, avoiding exposure in the frontend, and structuring backend requests to safely handle external API calls. We also gained experience debugging API-related issues and understanding rate limits and request flows.
User Interface & User Experience Design We developed a deeper understanding of how important UI/UX is in shaping user interaction. We focused on creating a clean, intuitive interface that guides users through the interview process from setup to completion, without confusion. We also learned how small design choices (layout, spacing, feedback, and flow) significantly impact usability and engagement.
Designing for Real User Interaction Beyond just visuals, we learned how to design with the user in mind anticipating how users would respond in a live interview setting. This included structuring screens logically, minimizing friction, and ensuring the experience feels natural and responsive, especially in a voice-based environment.

What's next for Ace.ai

Hybrid Interview Experience (Behavioral + Technical) We plan to integrate a hybrid interview mode where users are evaluated on both behavioral and technical skills within a single session. This creates a more realistic interview experience that mirrors real-world hiring processes.

Expanded Role-Specific Tracks We aim to introduce additional tech roles such as frontend, backend, data science, cybersecurity, and machine learning, allowing users to practice interviews tailored to their specific career path.
Resume-Based Personalization Future iterations will allow users to upload their resume so the AI can tailor questions and feedback based on their experience and skill set.