For a better experience on Michael Wills Update Your Browser

Indeed

CASE STUDY

CLIENT

Indeed Flex

TARGET

Clients and job seekers

Indeed is a global job search platform that connects job seekers with employers. It aggregates listings from thousands of sources and offers tools for applying, posting jobs, and managing recruitment, making hiring faster and more accessible worldwide.

My role was working across many different value streams, although this case study focuses on delivering an AI interview service to speed up recruitment.

AIMobileAndroid/IOSResearchDesign System

50,000+

Jobs created

500+

AI interviews conducted

1000+

Clients

My Role

Delivered new features across web and mobile platforms, turning high-level briefs into user-focused solutions through research and visual design.

Designed an AI-powered interview feature to streamline candidate screening and improve hiring efficiency.

Played a key role in evolving and maintaining a large design system for consistency and scalability.

Collaborated closely with developers to ensure components were visually aligned, technically feasible, and integrated with the development library.

Problem statement

Recruitment often involves lengthy screening and multiple interview stages, causing delays for both employers and candidates. To address this, we introduced an AI-powered interview feature that automates initial assessments. By conducting structured interviews and evaluating responses with AI, the solution reduces time-to-hire, improves screening consistency, and frees recruiters to focus on higher-value tasks.

Collaborated with product managers and machine learning engineers to develop an AI model for a candidate interview tool.

Worked closely with job sector leads and hiring managers to curate a comprehensive set of interview questions for integration into the AI model.

Designed and delivered the AI-powered interview feature, enabling candidates to complete interviews and employers to review results seamlessly.

Interviewers and employers to review results seamlessly. Ensured all components were designed and implemented in alignment with the established design system.

Requirements

DEFINING THE INTERVIEW QUESTIONS

I began by partnering with job sector leads and hiring managers to compile a well-structured set of interview questions, which served as the foundation for the AI model.

BUILDING THE AI INTERVIEW REQUISITION TOOL

I designed an AI-powered interview requisition tool that enables recruiters to easily select from a curated set of questions and configure dynamic outcomes tailored to each role.

Competitor benchmarking

I conducted in-depth competitor and market research on existing AI interview platforms and tools to analyze their features, user experience, and technological approaches. This process provided valuable insights into industry standards, emerging trends, and gaps in the market, helping shape a solution that was both innovative and aligned with recruiter needs.

The process

The diagram below illustrates the user flow, comparing a traditional human-led interview with an AI-driven interview, along with the potential NLP-based services we explored for integration.

Deliverables

I explored multiple design approaches to bring an AI voice agent to life, focusing on creating a natural and human-like conversational experience. This involved experimenting with various shapes, patterns, and UI layouts to establish visual cues that complemented the flow of dialogue. By iterating on different interaction models and interface elements, I aimed to strike the right balance between functionality and familiarity, ensuring the voice agent felt approachable and intuitive while maintaining a sense of personality.

The journey below illustrates the complete flow a candidate follows before starting the AI interview. Mapping this journey was critical to ensure clarity and preparedness, as all interviews are unmoderated. While the process is automated, it was essential to design steps that guide candidates effectively, provide the right context, and build confidence before they engage with the AI-driven experience.

The final steps focused on verifying that the candidate’s camera and microphone were functioning correctly to ensure a smooth interview experience. The AI interview combined on-screen written questions with audio prompts delivered through an AI-generated voice, creating a more natural and engaging interaction. While all questions were predetermined, the system was designed to dynamically adapt based on the candidate’s responses, allowing for a personalised and context-aware interview flow.

Structured feedback

The final stage involved designing and building a system that allowed recruiters to review AI interview results and make informed hiring decisions. This intelligent tool leveraged the AI model’s analysis, applying predefined rules and dynamic outcomes to recommend the highest-quality candidates. Recruiters were provided with a suite of features and decision-making tools, carefully defined through research and consultations with sector leads, ensuring the solution aligned with real-world hiring needs.

Previous project Walgreens Boots Alliance Next project Schneider Electric

Find out more

Got an interesting project and would like to work on it with me?

Get in touch Download CV