Elliot: the future of recruitment

From take home to home run

Timeline: 1 week

Check out the 10 minute video walkthrough, or the 1 minute summary below.

For a more details, scroll down the page or contact me using the magic orange button.

 

summary

 

A take home task that turned into a passion project

When I received a take home task to create a conversational AI I decided to go above and beyond, and treat it like a live client project. I started by reaching out to recruiters and hiring managers in my own network to understand what really slows down the hiring process. These insights and constant iteration shaped everything that followed. I designed a conversation flow that felt human, trustworthy, and useful. Only then was I ready to embody it as a working voice-based AI assistant named Elliot. Along the way, I discovered edge cases, created warmth and nuance and tackled call screening services. I tested it live with real users, iterated constantly, and ended up connecting with a C-suite leader at a fast-growing company who’s now interested in adopting the tool. In just a few days, I went from concept to conversation, simulation to solution — and uncovered a real business opportunity in the process.

 

Full story below. Let’s do this?

problem

 

Starting With People, Not Prompts

Before designing anything, I started where the real problems live — in the day-to-day reality of recruiters and hiring managers. I reached out to people in my network and asked a simple question:

“What’s actually been slowing down your hiring process?”

I wanted firsthand frustration. What I uncovered wasn’t surprising, it was very human:

  • Scheduling is a mess, especially across time zones

  • Candidates drop off when there’s silence or lag

  • Recruitment teams are small, but their task load is massive

Delays caused by scheduling ping-pong, timezone mismatches or lack of recruiter bandwidth are costing companies great candidates

This validated the opportunity for intervention and became the foundation an AI voice agent designed to speed up early-stage candidate screening.

ideation

 

Designing Conversations That Don’t Feel Robotic

Once I understood the stakes, I turned to the brief. Before touching any tools, I sketched out the conversation flow, not as a chatbot script, but as an exploration of organically branching human dialogue.

Talking to myself would only get me so far however. To really explore where this conversation could go, I had to get out of the room and get out of my own head.

I roleplayed the conversation with real users and unsurprisingly discovered a lot of frictions and deviations I had not thought of myself. No amount of whiteboarding beats user testing.

Pivotal insights:

Testing round 1:

  • Users apply for a lot of jobs - add a summary reminder at the start of the call.

  • The AI may catch them at a time they are not available.

  • Candidates are anxious for an explicit confirmation of the next interview.

Testing round 2:

  • Asking for the name first raises phishing suspicions.

  • Confident candidates may lead the conversation - the AI has to stay on target.

Testing round 3:

  • Candidates may have accessibility requirements.

  • Some users may be curious or put off by talking to an AI.

Each iteration added emotional intelligence to the interaction. This still wasn’t about building a AI. I was creating blueprint for the real human experience in this context.

 “How might we help businesses and job seekers move the recruitment process forward in a way that sounds natural?”

 
 
 

prototyping

Get some Moxy

Moxy is a mobile app that allows users to study, practice and get feedback on interview techniques and answers. At its core lies a database of interview questions, organised by industry and position. Each question is rated by how frequently it comes up in interviews, based on user feedback.

Users will choose the industry they are interested in during onboarding. The search functionality assumes no pre-existing knowledge of the field and allows the user to browse common job titles and the possible interview questions associated with them.

Selecting a questions gives access to curated model answers and technique tips, including transcripts for situations when reading is preferable. The user can also see answers published by other peers, encouraging feedback and horizontal mentoring.

All questions and videos can be bookmarked and reviewed in a separate section of the app. The bookmarked questions can then be treated as a playlist for the practice sessions, alongside question lists curated by job title.

Once recorded, the user can choose to save the video to their device or publish them to the community to receive feedback from others.

Marketing and next steps

I created a responsive marketing website mockup that fits into the branding and tone of the app, tackling acute pains in a users career with simplicity, softness and a degree of playfulness. Having been inspired by my research since the projects completion, I am considering iterating on the landing page, and a live launch to validate the idea on real users.

In the time of the Great Resignation, I believe that from a product perspective, the timing could be perfect to further research, develop and launch a suite of tools that will help people make sense of the changing world of work.

Still here?

Thanks or making it this far.

I’d love to know what you thought. Get in touch using the big orange button.