Lonesyte

An enterprise cultural-intelligence platform connecting global businesses with vetted local experts to de-risk cross-border decisions.

Local Support: a 3D globe with contributor counts clustered over the regions they cover
The business dashboard, with an onboarding checklist and quick actions for an organisation
The task board, counting open, in-review and closed tasks against an average turnaround
The Post Task dialog, collecting objective, output type, region, deadline, language and budget
The research surface, where a question becomes a study fielded to vetted contributors
Category
Enterprise Platform
Role
Full-stack engineer
Year
2025

Technology

  • NestJS
  • Prisma
  • Next.js
  • PostgreSQL
01.

The Problem

Companies expanding into unfamiliar markets make product, marketing, HR and partnership decisions on assumptions that do not survive contact with local culture. The expertise to correct those assumptions exists, but it is scattered and unvetted.

Connecting the two sides is not a directory problem. It requires matching the right expert to a specific business question, keeping competing organisations apart, and protecting confidential briefs.

02.

What We Built

Three surfaces over one backend: a marketing site, a contributor portal and a business dashboard.

The contributor lifecycle runs end to end: application intake, admin review, activation, profile, an earnings ledger and in-app notifications driven by deadline crons.

The business dashboard is role-gated and carries metrics, reviews and task templates. A shared task-creation surface feeds the matching engine, which applies rules-based matching with race-safe self-accept so two contributors cannot claim the same task, and NDA-gated deliverables so confidential material is only released to accepted parties.

03.

A Brief Is Structured Before It Is Sent

A task is not free text. Posting one collects the objective, the output type expected back, the region, the language, the deadline and the budget, so the matching engine has something to match on and the contributor knows what "done" means before accepting.

The same structure is what makes the work add up: reporting aggregates completed tasks by output type and risk type, and a completed task can be saved as a template so its brief and risk framing are reused rather than rewritten.

04.

Finding The Right Person

Contributors are found two ways. The directory filters on market, expertise, language, availability and rating; Local Support answers the other question — who is near a place — on a 3D globe that clusters contributors and companies by region and zooms into individual coverage.

Alongside one-to-one tasks, a research surface fields a question to many contributors at once as a survey or panel, so answers arrive as structured, comparable data rather than as a pile of separate deliverables.

06.

Outcome

The platform runs in production at lonesyte.com, serving contributors and business customers across both portals.

Other projects

Say hi, and let’s get to work

Let’s discuss your ideas

  • Construction Company
  • Mandi System
  • Build AI
  • Lonesyte