Project detail
Project detail
Project detail
One-Click Apply
One-Click Apply
One-Click Apply
Shaping AI behavior and interaction models to reduce cognitive load in high-volume job applications
Applying to multiple jobs shouldn’t feel like filling the same form again and again.
Applying to multiple jobs shouldn’t feel like filling the same form again and again.
Applying to multiple jobs shouldn’t feel like filling the same form again and again.
Problem
Problem
Applying to multiple jobs is mentally exhausting. Users are forced to re-enter the same information, make repeated decisions, and stay attentive across long, form-heavy flows. Over time, this creates cognitive fatigue, frustration, and drop-off — especially for frontline workers with limited time and digital patience.
Applying to multiple jobs is mentally exhausting. Users are forced to re-enter the same information, make repeated decisions, and stay attentive across long, form-heavy flows. Over time, this creates cognitive fatigue, frustration, and drop-off — especially for frontline workers with limited time and digital patience.
Applying to multiple jobs is mentally exhausting. Users are forced to re-enter the same information, make repeated decisions, and stay attentive across long, form-heavy flows. Over time, this creates cognitive fatigue, frustration, and drop-off — especially for frontline workers with limited time and digital patience.
~10 minutes per application
~10 mins/application
Workers didn’t need another form. They needed a system that could understand their intent, reduce cognitive effort, and handle repetition on their behalf.
Workers didn’t need another form. They needed a system that could understand their intent, reduce cognitive effort, and handle repetition on their behalf.
Workers didn’t need another form. They needed a system that could understand their intent, reduce cognitive effort, and handle repetition on their behalf.
Goal
Goal
Design an AI-driven application flow that removes repetition without removing user control.
Design an AI-driven application flow that removes repetition without removing user control.
Design an AI-driven application flow that removes repetition without removing user control.
Process
Process
Process
Research & Competitive Benchmark
Research & Competitive Benchmark
We analyzed application experiences across major hiring platforms to understand patterns of repetition, decision fatigue, and automation behavior — focusing on how users repeat actions and how little context carries forward between applications.
We analyzed application experiences across major hiring platforms to understand patterns of repetition, decision fatigue, and automation behavior — focusing on how users repeat actions and how little context carries forward between applications.
User Research – Key Behavioral Patterns
User Research – Key Behavioral Patterns
Through interviews and observation, three dominant behavior patterns emerged:
Through interviews and observation, three dominant behavior patterns emerged:
Through interviews and observation, three dominant behavior patterns emerged:
Key Insight
Across all patterns, users wanted help — not replacement.
They expected the system to reduce effort and repetition while staying visible, predictable, and controllable.
This directly shaped the interaction model toward assistive, human-in-the-loop automation, rather than full replacement.
Across all patterns, users wanted help — not replacement.
They expected the system to reduce effort and repetition while staying visible, predictable, and controllable.
This directly shaped the interaction model toward assistive, human-in-the-loop automation, rather than full replacement.
Across all patterns, users wanted help — not replacement.
They expected the system to reduce effort and repetition while staying visible, predictable, and controllable.
This directly shaped the interaction model toward assistive, human-in-the-loop automation, rather than full replacement.
Competitive Benchmark – Hiring Platforms
Competitive Benchmark – Hiring Platforms
We reviewed application flows across multiple platforms to understand how repetition, decision-making, and automation are handled.
Across platforms, the same structural patterns emerged.
We reviewed application flows across multiple platforms to understand how repetition, decision-making, and automation are handled.
Across platforms, the same structural patterns emerged.
We reviewed application flows across multiple platforms to understand how repetition, decision-making, and automation are handled.
Across platforms, the same structural patterns emerged.
Repetition is systemic
Users re-enter the same information (name, experience, availability, certifications) for every job. Context is rarely reused.
Repetition is systemic
Users re-enter the same information (name, experience, availability, certifications) for every job. Context is rarely reused.
Repetition is systemic
Users re-enter the same information (name, experience, availability, certifications) for every job. Context is rarely reused.
Decision fatigue is unaddressed
Users answer the same screening questions and confirm the same preferences again and again.
Decision fatigue is unaddressed
Users answer the same screening questions and confirm the same preferences again and again.
Decision fatigue is unaddressed
Users answer the same screening questions and confirm the same preferences again and again.
No intent recognition
Platforms do not adapt when users apply to multiple jobs. Each application is treated as isolated.
No intent recognition
Platforms do not adapt when users apply to multiple jobs. Each application is treated as isolated.
No intent recognition
Platforms do not adapt when users apply to multiple jobs. Each application is treated as isolated.
Automation without visibility
Basic autofill exists, but users are not told why data is filled or what the system is doing.
Automation without visibility
Basic autofill exists, but users are not told why data is filled or what the system is doing.
Automation without visibility
Basic autofill exists, but users are not told why data is filled or what the system is doing.
Key Takeaway
Design an AI-driven flow that recognizes intent, removes repetition, and assists with decisions — while keeping users in control.
Design an AI-driven flow that recognizes intent, removes repetition, and assists with decisions — while keeping users in control.
Design an AI-driven flow that recognizes intent, removes repetition, and assists with decisions — while keeping users in control.
Discovery & Insights
Discovery & Insights
Through user interviews, flow analysis, and drop-off data, we identified that the biggest friction in job applications wasn’t form length — it was repetition, decision fatigue, and uncertainty about what the system was doing on their behalf. Users were not struggling with understanding questions, but with having to answer the same ones again and again across different jobs.
Through user interviews, flow analysis, and drop-off data, we identified that the biggest friction in job applications wasn’t form length — it was repetition, decision fatigue, and uncertainty about what the system was doing on their behalf. Users were not struggling with understanding questions, but with having to answer the same ones again and again across different jobs.
Through user interviews, flow analysis, and drop-off data, we identified that the biggest friction in job applications wasn’t form length — it was repetition, decision fatigue, and uncertainty about what the system was doing on their behalf. Users were not struggling with understanding questions, but with having to answer the same ones again and again across different jobs.
1. Repetition creates cognitive fatigue
Users were forced to re-enter the same information across every application and make the same decisions repeatedly. Over time, this repetition created mental fatigue, frustration, and disengagement — even when users were highly motivated to apply.
2. Systems ignored clear user intent
We observed that users often had a clear intent — “I want to apply to many jobs quickly” — but the system forced them into slow, linear, form-heavy flows that ignored that intent.
3. Uncertainty reduced trust in automation
Users were uncomfortable when they didn’t understand what the system was doing on their behalf. The lack of visibility into automated actions and decisions made people hesitant to rely on the platform, even when it could save time.
Key Insight
The opportunity wasn’t to make forms shorter. It was to recognize intent, reduce cognitive effort, and let the system handle repetition — without removing user understanding, confidence, or control.
The opportunity wasn’t to make forms shorter. It was to recognize intent, reduce cognitive effort, and let the system handle repetition — without removing user understanding, confidence, or control.
The opportunity wasn’t to make forms shorter. It was to recognize intent, reduce cognitive effort, and let the system handle repetition — without removing user understanding, confidence, or control.
Exploring Interaction Models
Exploring Interaction Models
We explored multiple ways users and AI could collaborate during the application process — from fully manual flows, to aggressive automation, to hybrid models. The core question was not “how fast can we make this?” but “where should AI act, and where should the user stay involved?”
We explored multiple ways users and AI could collaborate during the application process — from fully manual flows, to aggressive automation, to hybrid models. The core question was not “how fast can we make this?” but “where should AI act, and where should the user stay involved?”
We explored multiple ways users and AI could collaborate during the application process — from fully manual flows, to aggressive automation, to hybrid models. The core question was not “how fast can we make this?” but “where should AI act, and where should the user stay involved?”
Early concepts included:
Fully automated bulk apply — AI submits applications with minimal user input
Step-by-step assisted flows — AI supports each field individually
Hybrid models — AI handles repetition and preparation while users review and confirm
Early concepts included:
Fully automated bulk apply — AI submits applications with minimal user input
Step-by-step assisted flows — AI supports each field individually
Hybrid models — AI handles repetition and preparation while users review and confirm
Early concepts included:
Fully automated bulk apply — AI submits applications with minimal user input
Step-by-step assisted flows — AI supports each field individually
Hybrid models — AI handles repetition and preparation while users review and confirm
Through iteration and testing, it became clear that trust and control mattered more than raw speed. Users were willing to let AI help — as long as they could understand what it was doing and override it when needed.
Through iteration and testing, it became clear that trust and control mattered more than raw speed. Users were willing to let AI help — as long as they could understand what it was doing and override it when needed.
Through iteration and testing, it became clear that trust and control mattered more than raw speed. Users were willing to let AI help — as long as they could understand what it was doing and override it when needed.
This led us to a human-in-the-loop interaction model, where AI progressively takes on repetitive work while users retain visibility and final control.
This led us to a human-in-the-loop interaction model, where AI progressively takes on repetitive work while users retain visibility and final control.
This led us to a human-in-the-loop interaction model, where AI progressively takes on repetitive work while users retain visibility and final control.
Interaction Model & Experience Design
Interaction Model & Experience Design
One Click Apply is built on a clear interaction model that defines how AI learns, assists, and hands control back to the user.
Each behavior is intentionally expressed through the interface so users always understand what the system is doing and why.
One Click Apply is built on a clear interaction model that defines how AI learns, assists, and hands control back to the user.
Each behavior is intentionally expressed through the interface so users always understand what the system is doing and why.
How do we make AI feel helpful without making it feel invisible or overwhelming?
How do we make AI feel helpful without making it feel invisible or overwhelming?
How do we make AI feel helpful without making it feel invisible or overwhelming?
Learn Once
System behavior
The system captures user information a single time through resume parsing or manual input, along with baseline questions required across employers.
The system captures user information a single time through resume parsing or manual input, along with baseline questions required across employers.
How it appears in the UI
Profile setup is lightweight and progressive. Parsed resume data is previewed and fully editable, reinforcing accuracy and trust from the first interaction.
Profile setup is lightweight and progressive. Parsed resume data is previewed and fully editable, reinforcing accuracy and trust from the first interaction.
Reuse Context
System behavior
When users apply to multiple jobs, previously captured information and answered questions are automatically reused. The same input is never requested twice.
When users apply to multiple jobs, previously captured information and answered questions are automatically reused. The same input is never requested twice.
How it appears in the UI
Long application forms are replaced with a short confirmation view. Users review what will be applied instead of re-entering data.
Long application forms are replaced with a short confirmation view. Users review what will be applied instead of re-entering data.
Ask Only When Unknown
System behavior
If a specific job requires information that hasn’t been captured before, the system asks only that question and stores it for future use.
If a specific job requires information that hasn’t been captured before, the system asks only that question and stores it for future use.
How it appears in the UI
New questions appear inline at the moment they are needed, clearly marked as being saved for future applications.
New questions appear inline at the moment they are needed, clearly marked as being saved for future applications.
Provide Clear Feedback
System behavior
After submission, the system communicates completion and next steps explicitly.
After submission, the system communicates completion and next steps explicitly.
How it appears in the UI
A clear success state confirms what was submitted, where it was sent, and what happens next — avoiding silent or ambiguous transitions.
A clear success state confirms what was submitted, where it was sent, and what happens next — avoiding silent or ambiguous transitions.
Respect Real-World Boundaries
System behavior
When a step requires human involvement (interviews, assessments, verification), automation stops.
When a step requires human involvement (interviews, assessments, verification), automation stops.
How it appears in the UI
The interface clearly signals the transition from AI assistance to human interaction, maintaining a consistent mental model.
The interface clearly signals the transition from AI assistance to human interaction, maintaining a consistent mental model.
The visual design is intentionally calm and restrained.
Hierarchy, spacing, and motion are used to reduce cognitive load and reinforce trust — ensuring AI feels assistive, predictable, and human.
The visual design is intentionally calm and restrained.
Hierarchy, spacing, and motion are used to reduce cognitive load and reinforce trust — ensuring AI feels assistive, predictable, and human.
Impact
Impact
Impact
Impact & Outcomes
Impact & Outcomes
Design
Design
Design
Design System & Brand Expression
Design System & Brand Expression
Alongside interaction design, I was responsible for shaping the visual language and component system to ensure consistency, clarity, and scalability.
Alongside interaction design, I was responsible for shaping the visual language and component system to ensure consistency, clarity, and scalability.
Alongside interaction design, I was responsible for shaping the visual language and component system to ensure consistency, clarity, and scalability.
My Role & Scope
My Role & Scope
Product Lead
Product Lead
Product Lead
6 months
6 months
6 months
Circle Global
Circle Global
Circle Global
End-to-end experience design · interaction model · design system
End-to-end experience design · interaction model · design system
End-to-end experience design · interaction model · design system
Contact
Contact
Contact
Let's Get in Touch
Interested in collaborating or learning more? Feel free to reach out.