InterView

Example
Statistics
InterView
Tech Stack Diagram

Inspiration

The spark that ignited "InterView" was our collective realization of the pervasive issue of unconscious biases in recruitment processes. With the job market more saturated than ever, we recognized the critical need for tools that not only help candidates but also empower recruiters to refine their interviewing skills. Inspired by the vision of equitable hiring practices, we set out to develop a tool that could address these biases head-on, ensuring every candidate has a fair chance based on their skills and potential.

What it does

"InterView" is an innovative AI-powered website that helps HR departments train new recruiters by providing them with immediate feedback on their interviewing techniques. By processing interview audio in real-time, our tool identifies potential biases related to age, gender, or ethnic background. It then offers constructive feedback, guiding recruiters on how to conduct more equitable and effective interviews.

How we built it

We built "InterView" using a combination of cutting-edge technologies and frameworks. The frontend is crafted with React, creating an intuitive and responsive user experience. For audio processing, we integrated Google's speech-to-text API to transcribe interviews accurately. The core of our bias detection mechanism utilizes Google Gemini, Google Vertex AI, and Langchain for RAG. Data from the job posting and many examples of bias and unjust interview questions are embedded using Vertex AI Embeddings and then stored in a Google BigQuery database. Then, when an interviewer asks a question, RAG is used to find relevant key information from the vector store and that information is used alongside Gemini's existing knowledge base to create tailored responses that provide strong analysis of the question they posed.

Challenges we ran into

One of the major challenges we faced was ensuring the accuracy and sensitivity of our bias detection. Reducing the number of undetected biases required extensive testing and fine-tuning. Another challenge was processing audio in real-time with minimal latency, critical for providing immediate feedback. Navigating these technical hurdles while maintaining a seamless user experience was a learning curve for the team. We overcame this solution by implementing a pipeline using a Redis Cloud database where two servers work concurrently. One server (in FastAPI) runs the transcription and RAG whereas the other server (in Flask) continuously retrieves content from the database. Thanks to the concurrent nature of the operations, there is minimal delay between transcription, processing, and returning data back to the frontend to be displayed.

Accomplishments that we're proud of

We are incredibly proud of developing a tool that bridges a significant gap in the recruitment industry. Successfully integrating complex AI technologies to create a user-friendly platform that can genuinely impact hiring practices is an achievement that resonates with our core values. Additionally, our ability to work as a cohesive team, leveraging each member's strengths to overcome challenges, stands as a testament to our dedication and collaborative spirit.

What we learned

Throughout this journey, we've gained invaluable insights into AI's potential to address societal challenges, specifically in creating more inclusive workplaces. We learned the intricacies of processing natural language and speech recognition, and the importance of designing with empathy, ensuring our tool genuinely serves the needs of recruiters and candidates alike. This project has also reinforced the value of teamwork, perseverance, and the relentless pursuit of innovation.

What's next for InterView

The future of "InterView" is bright and bustling with possibilities. We plan to refine our bias detection algorithms further, incorporating more nuanced understanding of language and cultural differences. Expanding our tool to support multiple languages and integrating it with existing HR platforms are immediate next steps. Long-term, we envision "InterView" becoming an indispensable tool for companies worldwide, leading the charge towards truly inclusive and unbiased hiring practices.