Nominait is a dual-sided HR platform designed to improve hiring outcomes through AI-assisted profiles, transparent match scoring, and structured interviews. The product supports two distinct journeys-recruiters searching and contacting candidates, and candidates onboarding into an AI-powered profile and interview flow that feeds the matching engine. I designed the end-to-end experience across onboarding, dashboards, job matches, recruiter outreach, and the monetization layer built around tokenized contact unlocks.
Context
Nominait targets mid-to-large organizations in Saudi Arabia where hiring is often slow, manual, and compliance-heavy. Workflows typically involve multiple stakeholders (CHRO/HRD, TA leads, hiring managers, ops), and hiring decisions require both speed and auditability.
The problem
Enterprise hiring repeatedly breaks in the same places:
High setup friction: creating company profiles and job requirements is time-consuming.
Low-quality signals: job descriptions and candidate inputs are inconsistent, weakening matching.
Trust barrier with AI: teams need control, reversibility, and clear explanations-not a black box.
Success criteria and North Star
When I didn’t have reliable post-launch metrics available, I defined success in terms of observable adoption and decision confidence.
North Star (product): Time to validated match - time from first entry to the moment a user can confidently decide: shortlist / contact / save / proceed.
Leading indicators:
Recruiter: onboarding completion, % roles published, shortlist actions, match deep-dives opened, contact unlocks.
Candidate: profile completeness, AI interview completion, saved jobs, match deep-dives opened.
Design principles
Time-to-first-value: get users to “ready-to-match” quickly.
AI with user control: AI suggests; users approve, edit, or revert.
Explainability by design: show why a score exists and how to act on it.
Enterprise reality: collaboration + role separation + auditable flows.
The story
Chapter A - Recruiter activation
Goal
Reduce setup friction and get the company to a usable profile fast, while building trust in AI through preview, reversible actions, and manual overrides.
What I designed
A guided onboarding flow with clear progress and role selection.
Company profile creation via AI-assisted draft + live preview.
A reusable AI pattern: suggest → preview → approve → undo/redo → manual edit.
An enterprise-friendly option to skip job creation (setup and recruiting often have different owners).
Why this matters
Enterprise users adopt tools when they can reach value quickly and maintain control. In this flow, AI reduces work, but the UI makes changes visible, auditable, and reversible.
Sign-in
Role selection
Onboarding overview
Company profile creation 1/2
Company profile creation 2/2
Company Profile Agent
Role Definition Agent 1/2
Role Definition Agent 1/2
AI-assisted company setup (reduce time-to-first-value) We designed onboarding to help enterprise users reach a “ready-to-match” state fast, without long forms upfront. The flow starts with role selection (Hiring vs. Job seeker) to personalize the experience from the first click.
For company setup, we generate a first draft by extracting publicly available information and translating it into a structured company profile. To build trust and maintain control, the UI pairs an AI “Profile Agent” panel with a live profile preview, plus reversible edits (undo/redo) and a manual edit modal as an escape hatch.
We also introduced a “skip job creation” option to match real enterprise workflows where account setup and job posting are often owned by different people.
Chapter B - Recruiter matching & review
Goal
Enable recruiters to review candidates quickly while staying data-driven and confident in AI recommendations.
What I designed
A job listings operational hub (Active vs Draft, and quality gates like “Needs review”).
Candidate review per role with a clear Match Score and quick shortlist actions.
A two-level explainability model:
Inline expansion for quick scanning.
Deep dive modal for high-confidence decisions and stakeholder alignment.
Integrated documents access to reduce back-and-forth and support compliance-heavy processes.
Company dashboard
Job Listings (All / Active / Draft + Needs review)
Matched candidates listing
Expanded candidate card
Candidate documents
Matching & review: explainable shortlists for data-driven recruiters After onboarding, the product shifts into its core workflow: reviewing candidates per role. Recruiters are outcome-driven and highly data-oriented, so we designed matching around explainability and decision confidence, not just ranking.
The job listings area provides an operational view (Active vs Draft, quality states) with clear actions to edit, publish, and open candidate pools. Inside each role, candidates are presented with a visible Match Score and a lightweight “save/shortlist” action to build a working set.
To avoid a black-box experience, we introduced a two-level explanation model:
Inline insights for rapid scanning (AI summary, match breakdown, and improvement areas)
A full ‘Job Match’ view for deep dives, stakeholder alignment, and confident outreach
We also added “Interview focus” hints to translate model outputs into practical next steps and reduced recruiter context switching by keeping documents within the same flow.
Chapter C - Candidate activation
Goal
Create a high-signal candidate profile with minimal friction (LinkedIn/CV import), then enrich it with AI-assisted refinement and a short interview to improve match quality beyond keywords.
What I designed
LinkedIn sign-in as a fast lane.
Guided onboarding with clear expectations and step framing.
Candidate Profile Agent (split view) with task-based CTAs instead of a blank chat.
Manual edit as an escape hatch.
AI Interview positioned as a differentiator: behavioral signal collection + confidence calibration.
LinkedIn sign-in as a fast lane.
Guided onboarding with clear expectations and step framing.
LinkedIn imported review
Resume parsing
Candidate Profile Agent (split view) + improvement CTAs
Candidate Profile Agent (manual edit modal)
AI interview prep
AI interview question screen
Candidate onboarding: high-signal profiles with LinkedIn import, AI refinement, and an AI interview To improve matching quality, we designed candidate onboarding as a structured way to capture high-signal inputs with minimal friction. Candidates can sign in with LinkedIn to import identity and work history, then upload a resume to enrich profile data.
Instead of forcing long forms, we use an AI-assisted Profile Agent to refine the summary and skills while providing a live profile preview. The interface is task-based (“Improve summary”, “Add missing skills”, “Rewrite profile”) to prevent the blank-chat problem and give candidates clear next actions.
To go beyond keyword matching, onboarding includes a short AI interview with behavioral questions to capture communication and work-style signals. We intentionally placed the interview after profile generation so candidates understand the purpose and feel prepared. Microphone permissions are requested explicitly, with clear guidance and the option to skip and complete later.
Chapter D - Candidate job matches
Goal
Help candidates understand why a role is a good fit, what to improve, and how to prepare—turning matching into an actionable coach rather than a ranked list.
What I designed
A candidate dashboard that reinforces the quality loop: improve profile → better matches.
Job Matches list with visible Match Score and fast scan patterns.
Inline expansion + modal deep dive for explainability.
Actionability layer: strengths, growth opportunities, and interview tips.
Candidate dashboard
Job Matches list (All / Saved)
Overview breakdown (strengths, growth opportunities)
Job Description tab
About Company tab
Candidate experience: explainable matches that translate AI signals into next steps After onboarding, candidates land on a dashboard that reinforces the product loop: improve profile quality → get better matches → validate fit → prepare for interviews. We intentionally position “Improve your profile” as a prominent action, because profile completeness directly impacts match accuracy.
In Job Matches, each role includes a clear Match Score and an expand/collapse pattern that reveals an AI-driven summary and a structured breakdown. The experience explains fit across multiple dimensions (technical fit, culture/personality), shows supporting evidence (skills, experience, preferences), and highlights both strengths and growth opportunities.
We also added interview-focused guidance to help candidates act on insights immediately: what to emphasize, what to prepare, and which skills to improve. For deeper evaluation, the match view expands into a full modal deep dive with visualizations and the same tab structure-keeping scanning fast while supporting high-trust decisions when it matters.
Monetization (bonus) - Tokenized contact unlocks (Free trial → Pro)
Goal
Introduce freemium pricing without breaking evaluation: recruiters can review matches and build shortlists, but contacting candidates becomes the value moment tied to plan limits.
CTAs to upgrade throughout the web application
Unlocking contact 1/2
Unlocking contact 2/2
Plan upgrade CTA
Freemium monetization: tokenized contact unlocks at the moment of value To support a scalable go-to-market model, we introduced tokenized “contact unlocks” as a freemium mechanic. Recruiters can browse matches, review explainable scoring, and build shortlists, then unlock contact details when they’re ready to act.
The gating is transparent and contextual: remaining unlocks are visible on the dashboard and inside the match view. When a user reaches the limit, the experience escalates into a clear upgrade flow that communicates benefits without disrupting evaluation.
My contribution (what I owned)
Designed end-to-end flows for Recruiter and Candidate sides (activation → matching → review).
Defined a reusable AI-assisted editing pattern (suggest → preview → approve → undo/redo → manual edit).
Created a two-depth explainability model (scan inline → deep dive modal) used across recruiter and candidate experiences.
Built information architecture for job listings, candidate review, and match breakdown.
Shaped product quality loops: profile improvement → better matches → better decisions.
What I would improve next
Score calibration: clearer interpretation of high/medium/low with thresholds and examples.
End-to-end hiring pipeline: match → outreach → interview scheduling → offer (if in scope later).
Experimentation: A/B tests for onboarding step order and AI interview placement.
AI Interviews: Design Interviews Feed to pick different interviews matched to candidate's profile and job listings.
