Selected work

Real systems built across AI, healthcare, and education.

Focused on building products that are reliable, scalable, and used in real environments. Each project below is a system I designed and built end-to-end.

Start a Conversation

Case Study

VertBoard — Digital Signage & Ad Marketplace Platform

Full-scale SaaS platform for managing screens, scheduling content, and monetizing ad inventory

Problem

Businesses with digital screens (TVs, billboards, vehicles) struggle to:

  • Manage content across multiple locations
  • Schedule playlists efficiently
  • Monetize unused screen time

At the same time, advertisers lack access to a unified platform to run targeted campaigns on physical screens.

Solution

Built a full-stack platform that combines:

  • Digital signage management
  • Ad marketplace
  • Campaign targeting and analytics

The system allows screen owners to control content and generate revenue, while advertisers can run targeted campaigns across a distributed network.

What I Built

  • Multi-tenant SaaS platform
  • Playlist and scheduling engine
  • Ad slot monetization system
  • Campaign creation and targeting system
  • Real-time analytics dashboards
  • Wallet and transaction system
  • Location-based targeting system

Tech Stack

Frontend:

  • React / Next.js

Backend:

  • Node.js / NestJS

Database:

  • MySQL / Firestore

Infrastructure:

  • Firebase / AWS

Outcome

  • Unified platform for screen management and monetization
  • Enabled advertisers to run targeted campaigns
  • Scalable system supporting multiple locations and screens
  • Real-time analytics and reporting

Case Study

HudLink — AI-Powered Due Diligence Intelligence Platform

Private AI infrastructure for transforming financial data into decision-ready intelligence

Problem

Due diligence in private markets is:

  • Manual and time-consuming
  • Dependent on fragmented data rooms
  • Prone to human error
  • Limited by generic tools that lack domain intelligence

At the same time, using public LLMs introduces serious risks:

  • Data leakage
  • Lack of control
  • No ownership of intelligence

Solution

Built a private AI-powered platform that transforms raw data into structured, decision-ready intelligence.

The system combines:

  • AI-driven document understanding
  • Structured workflows
  • Proprietary intelligence models

All running in a controlled environment where:

  • No data is exposed to public LLMs
  • Firms retain full ownership of their data and models

What I Built

  • End-to-end due diligence platform
  • Data room ingestion and structuring system
  • AI-powered analysis pipelines
  • Risk detection and reporting engine
  • Custom workflow systems for deal teams
  • Scalable backend architecture for high-volume data

AI & Infrastructure

Built around Small Language Models (SLMs) for:

  • Better control
  • Lower latency
  • Domain-specific intelligence

Designed with data sovereignty as a core principle:

  • Zero reliance on public LLM APIs
  • Fully controlled AI environment

Integrated private model training pipelines:

  • Models adapt to firm-specific workflows
  • Compounding intelligence over time

Developed intelligence infrastructure layer (Compact Machines):

  • Foundation for building custom AI workflows
  • Extensible system beyond a single application

Tech Stack

Frontend:

  • React / Next.js

Backend:

  • Node.js / NestJS

AI Layer:

  • Small Language Models (SLMs)
  • Custom AI pipelines

Infrastructure:

  • Cloud + private compute environments

Database:

  • Scalable data systems

Outcome

  • Eliminated weeks of manual due diligence work
  • Enabled faster, more reliable decision-making
  • Maintained full data privacy and control
  • Created a scalable intelligence system for private markets

Case Study

BabyDay — Smart Baby Care & Health Tracking Platform

Mobile parenting companion that delivers trusted baby care guidance, practical tools, and doctor support in one app

Problem

Parents struggle to:

  • Track feeding, sleep, diaper changes, and symptom history consistently on mobile
  • Get trusted, contextual guidance without jumping across fragmented apps and blogs
  • Reach pediatric doctors quickly when they need support on urgent concerns
  • Manage routines, medication, and appointment reminders without missing critical events

Most solutions handle only one part of the journey, but parents need one reliable system that combines care tracking, expert guidance, and medical communication.

Solution

Built a mobile-first platform that allows parents to:

  • Track baby activity in real time with low-friction logging flows
  • Access trusted baby care guidance tailored to age and development stage
  • Connect with doctors through in-app support flows
  • Receive reminders and actionable care insights generated from activity patterns

The app unifies tracking, guidance, and doctor support into one continuous parenting experience.

What I Built

  • Cross-platform mobile experience for daily baby care workflows
  • Structured activity tracker for feeding, sleep, diaper, and wellness events
  • Guidance modules with medically reviewed tips and contextual recommendations
  • Doctor support and consultation workflows for real-time parent assistance
  • Reminder engine for routines, medication, and appointment adherence
  • Secure identity, subscription, and family profile management

Tech Stack

Frontend:

  • React / Next.js

Backend:

  • Node.js

Database:

  • Scalable data layer for activity timelines and health events

Infrastructure:

  • Cloud-based architecture with secure API and notification services

Outcome

  • Unified mobile companion for day-to-day baby care management
  • Higher parent confidence through trusted guidance and clearer activity visibility
  • Fast doctor support channel embedded directly in the app journey
  • Scalable consumer health-tech foundation ready for user growth
Education · EdTechAILA · Apr 2024 – Dec 2025

AILA Platform

Multi-platform education system

01

Problem

An education organization needed a unified system that worked consistently across web, desktop, and mobile — with different interfaces for students, teachers, and administrators, all backed by the same data layer.

Solution

Built a full product ecosystem with shared backend services and platform-specific frontends. Students access content on web and mobile; administrators manage everything from a desktop app. All platforms share a single API layer and database, keeping data consistent across every surface.

What I Built

  • Web application for student learning and content delivery
  • Electron-based desktop application for administrators
  • React Native mobile app for students on iOS and Android
  • Shared NestJS backend with role-based access control
  • AI-driven study tools and automated academic support workflows
ReactElectronReact NativeNestJSMongoDBAWS

Outcomes

✓Single codebase approach reduced cross-platform inconsistencies
✓Unified backend made feature rollouts simultaneous across platforms
✓Student engagement metrics improved after mobile launch
Healthcare · ClinicalsiParadigm · Feb 2022 – Apr 2024

Digital Pathology Suite

Clinical workflow system for pathology operations

02

Problem

Pathology labs were running fragmented workflows across disconnected systems — sample tracking, image annotation, report generation, and case assignment were all handled separately, creating delays and errors in clinical operations.

Solution

Built a unified clinical workflow platform that connects sample intake, digital slide viewing, automated image processing, and report generation in a single system. Designed for high reliability and precision — errors in clinical software have direct consequences.

What I Built

  • High-reliability sample tracking and case management system
  • Digital slide viewer with annotation tools for pathologists
  • Automated image processing pipeline with AI-assisted analysis
  • Report generation workflows with structured output and audit trails
  • Role-based access for lab staff, pathologists, and administrators
ReactNode.jsMySQLDockerAWSWebSockets

Outcomes

✓Case turnaround time reduced significantly across lab operations
✓Eliminated manual handoffs between disconnected systems
✓System deployed in production without downtime incidents

What the interfaces look like

Clean, data-dense UIs built for real workflows.

Platform metrics

MRR

$128K

+22%

Users

3,941

+31%

Uptime

99.97%

Churn

1.1%

-0.4%

Monthly revenue

JFMAMJJASOND

Account list

CompanyMRRStatus
Acme Corp$4,200Active
TechFlow$890Active
Skyline$190Trial
NovaLabs$4,200Active

Sprint progress

Auth module100%
Dashboard UI78%
API layer55%
Reporting30%
Mobile views12%

System health

API Gateway
99.98%
Database cluster
99.95%
Worker queue
99.91%
File storage
100%
Email service
97.8%

User settings

Display name

Alex Mercer

Email

alex@company.com

Role

Admin
Save changes

What this work demonstrates

Consistent patterns across every project.

01

Building complex systems end-to-end

From data modelling to deployed product — I handle the full stack without handoff gaps.

02

Designing scalable architectures

Systems built to handle growth from day one. Not patched later when things break.

03

Creating high-performance applications

Fast load times, efficient data fetching, and UIs that hold up under real usage.

04

Handling real-world product challenges

Multi-platform systems, clinical-grade reliability, AI pipelines — products that have real consequences.

How this translates to your project

I can help you design and build systems like these — whether it's a SaaS platform, internal tool, or data-heavy dashboard. I work best on projects where there's a real product to build and real users who depend on it.

See how to get in touch →

Ready to build?

Need something like this built?

Tell me what you're building. I'll tell you clearly whether I can help and what that looks like.

Start a Conversation