Smeron
All Work
Career advancement platform

Candidate App

  • Mobile app
  • System design
  • Real-time sync
Candidate App — Career advancement platform

A mobile hub for career advancement, resume optimisation and specialised technical mentorship, bridging candidates and industry experts.

Who it was built for

The business, and what they came with.

Built for

The North Tech Labs

A career services company. They coach job seekers through resume and LinkedIn optimisation, mock interviews and technical mentorship, and sell that help as packages.

The brief

The business ran on people: a specialist, a candidate, and a long thread of email in between. We built the app that turns those services into something a candidate buys, works through and can see progress on, and that a mentor can deliver to many candidates at once.

What we delivered

What The North Tech Labs has now that they did not have before.

  • The whole app, both stores

    One React Native and Expo codebase released to iOS and Android through EAS. Tamagui holds the design tokens, so new services drop into a layout that already looks right.

  • Services turned into products

    Resume review, LinkedIn optimisation, mock interviews and mentorship each became a purchasable item with its own workflow. What used to be quoted over email now has a price and a buy button.

  • Dashboards and charts the candidate can read

    Profile strength, completed tasks, mentor feedback and what to do next, drawn rather than described. The candidate gets a reason to open the app mid-week; North Tech Labs gets a record that the service was delivered.

  • A mentor's working queue

    Tasks are assigned, submitted and reviewed in one place, with real-time chat and typing indicators. Specialists spend their hours reviewing work instead of chasing attachments.

  • Payment inside the app

    Stripe's React Native SDK with multi-currency support, so a candidate anywhere pays in their own currency at the moment they decide to.

Where the growth comes from

Why each piece of the build pays for itself.

  • Buying happens at the moment of intent

    A candidate decides their resume is the problem and pays on that screen. No quote, no invoice, no follow-up call that never gets returned.

  • One specialist covers more candidates

    Coordination was the cost that capped how many people North Tech Labs could take on. Structured tasks and a single review queue take it out of the mentor's day.

  • Visible progress sells the next package

    A dashboard that shows what is finished and what is missing is what turns a one-off resume review into mentorship.

The business analysis

Market, model and architecture. Every figure is either cited or labelled as a model.

The North Tech Labs already had a working business. What it did not have was a business that could grow, because every unit of revenue required a specialist's hours and a long email thread to coordinate them. The app is an attempt to convert a services company into a product company without changing what it sells — the same resume reviews and mock interviews, packaged so that a candidate can buy one at eleven at night and a mentor can deliver forty of them in a week.

Resume writing services market, 2025
$2.49bn
Forecast to reach $3.64bn by 2030 at a 7.9% CAGR. This is the narrow category the first purchasable item sits in.

Source · The Business Research Company, Resume Writing Service Global Market Report

Career coaching services market, 2025
$3.09bn
Forecast to $6.80bn by 2033 at 11.2% CAGR — growing faster than resume writing, and the higher-value package.

Source · Verified Market Research, Career Coaching Service Market

Services turned into products
4
Resume review, LinkedIn optimisation, mock interviews, technical mentorship — each with a price and a buy button.

Model · Counted from the delivered scope in this case study.

Codebase, two app stores
1
React Native and Expo released to iOS and Android through EAS. Two storefronts, one team.

Model · Architectural fact, from the build described in this case study.

Two markets, and the app moves customers between them

Career services split into a low-value, high-volume category and a high-value, low-volume one. Resume writing is the first: a $2.49bn market in 2025, forecast to $3.64bn by 2030 at a 7.9% compound rate 1. Coaching and mentorship is the second: $3.09bn in 2025, forecast to $6.80bn by 2033 at 11.2% 2. Both sit inside a recruitment technology market valued at $15.18bn in 2025 3.

The two categories North Tech Labs sells into, USD billions

Resume writing is the larger entry point today; coaching is the faster-growing and higher-value destination. The forecast columns are the research houses' own and are drawn hollow.

  • $2.49bn
  • $3.64bn
  • $3.09bn
  • $6.80bn
  • Resume 2025
  • Resume 2030
  • Coaching 2025
  • Coaching 2033

Hollow columns are the published forecast, not a measurement.

Source · The Business Research Company (resume writing) and Verified Market Research (career coaching)

The constraint was never demand, it was coordination

Every service North Tech Labs sold worked. The ceiling was that each one arrived as an email, was scoped in a reply, delivered as an attachment, and reviewed in another thread. Coordination cost is invisible in a services business because it is never invoiced — it just quietly determines how many candidates a specialist can carry.

Where a specialist's hour went before the app

Our reconstruction of the pre-app workflow described in this case study, apportioned by activity. It explains what the build removed; it is not a time-and-motion study.

Coordination and chasing35%
Scoping over email, chasing attachments, agreeing times, confirming what was already agreed.
Actual review and feedback40%
The only part the candidate is paying for.
Admin and record keeping15%
Tracking who bought what and whether it was delivered.
Payment and invoicing10%
Quoting, invoicing and following up on unpaid work.

Model · Our apportionment of the manual workflow this case study describes replacing. Directional, and not measured at the client.

Buying at the moment of intent

A candidate decides their resume is the problem at a specific moment, usually after a rejection. Under the old model that moment produced an enquiry, and the enquiry produced a quote, and the quote arrived after the feeling had passed. Stripe's React Native SDK with multi-currency support is in the build because the distance between deciding and paying is the single largest leak in a services funnel.

StepUnder the old modelIn the app
Scope agreedA reply, some hours laterStated on the product screen, before the decision
Price givenA quote, negotiatedFixed, visible, and the same for everyone
Payment takenAn invoice, then a chaseAt the moment of intent, in the candidate's own currency
Delivery trackedAn email thread nobody can auditA task list with uploads and deadlines
Progress shownAsserted in a callDrawn on a dashboard, and it points at the next purchase

What a candidate actually does

Five steps, each of which used to be an email. The offshoots are the states a services business normally handles with a phone call.

  1. Onboard

    Sign in, set a target role and a goal. The target role is what everything downstream is scored against.

  2. Buy a service

    Scope, price and delivery time are on the screen rather than in a reply. This is where the funnel used to end.

    • ↳ Multi-currency at checkout, so an overseas candidate pays in their own
  3. Work the tasks

    Each purchased service becomes a checklist with uploads and deadlines, which is what makes delivery auditable.

  4. Submission revieweddecision

    A specialist marks the work and answers in chat, with typing indicators and presence so the exchange feels live.

    • ↳ Needs rework → task reopened with feedback
    • ↳ Candidate goes quiet → deadline and reminder, not a chase email
  5. Progress moves, and points somewhere

    Profile strength and completion are drawn rather than described, and the gap in the chart is the next thing to buy.

The stack, by responsibility

A mobile product with a real-time layer bolted to a conventional REST core — the split matters, because chat and records have opposite requirements.

  1. Client

    One codebase for both stores, with design tokens so a new service drops into a layout that already looks right.

    • React Native
    • Expo SDK 54
    • TypeScript
    • Tamagui
  2. Client state

    Holds candidate, purchase and task state across a deep navigation tree without prop-drilling through it.

    • Zustand
  3. Records

    The durable half: purchases, tasks, submissions, feedback. Conventional REST, because these are conventional reads and writes.

    • Node.js
    • Express
    • REST
  4. Live layer

    Chat, typing indicators and notifications. Separated from records because a dropped message and a lost purchase are different severities.

    • Socket.io
  5. Commercial services

    Identity, payment and release. The parts that would be reckless to build.

    • Clerk
    • Stripe React Native SDK
    • EAS

What we would watch

RiskWhy it bitesEarly indicator
Specialist supply, not demandProductising the service makes it easy to sell more than the mentor bench can review. The queue becomes the product experienceReview turnaround lengthening while purchase volume is flat
Store dependency on paymentsIn-app purchase policy for digital services is contested and changes by platform; a ruling here moves gross margin directlyStore review friction on the purchase flow rather than on features
Progress as vanity metricA profile-strength score that does not correlate with getting hired is discovered eventually, and it is the thing selling the upgradeCandidates completing every task and disputing the outcome

References

  1. 1.Resume Writing Service Global Market Report · The Business Research Company
  2. 2.Career Coaching Service Market — size, growth, trends and forecast · Verified Market Research
  3. 3.Online Recruitment Technology Market · Fortune Business Insights

What a candidate does

The path a real user takes, in order.

  1. Onboard

    Sign in, set a target role and a goal.

  2. Buy a service

    Scope, price and delivery time are in the app, not in a reply.

  3. Work the tasks

    Each service becomes a checklist with uploads and deadlines.

  4. Get reviewed

    A specialist marks the submission and answers in chat.

  5. Watch it move

    Charts track profile strength and completion, and point at the next step.

Who it is for

The people whose problem this solves, and what they came for.

  • Job seekers

    Want a resume, a LinkedIn profile and interview answers fixed by someone who hires for a living.

  • Career switchers

    Need a structured path with checkpoints, not more advice.

  • Mentors and specialists

    Deliver reviews and sessions, and want the admin out of the way.

Reference

For the technical reader. Everyone else has what they need above.

Recommended case studies

The closest work to Candidate App, scored for relevance rather than picked by position.

Want the version of this built for your business?

Fifteen minutes with the people who shipped it. No deck, no account manager.