Software studio · est. 2026

We build software that's
actually finished.

UltiSoft is a two-person studio run by Sam Raths and Erik Arnold. We design, engineer, and ship real products — tested, secured, and shipped — across sports analytics, Mac tooling, and native apps.

Products shipped
5
Shipped & live
on Railway
Reach
Mac-native & web

Selected work

Products we're proud of

Each one is a real, shipping application — not a demo. Three on the web, two native on iOS and Mac. Different domains, one standard of craft.

UltiStats play library showing an animated ultimate frisbee playbook
Flagship Sports analytics

UltiStats

Real-time ultimate frisbee stats you run from the sideline.

A mobile-first PWA for tracking throws, catches, blocks, and scores on a live field, with instant line-performance analysis, team rosters, an animated playbook, and USA Ultimate tournament sync. One platform, three surfaces: the native app, a Google Sheets dashboard, and a computer-vision pipeline that pulls stats straight from game film.

  • Vanilla JS + Vite
  • PWA / Capacitor iOS
  • Express + SQLite
  • YOLO11 CV
Macinit app catalog for building a one-line macOS installer
Flagship Mac tooling

Macinit

Pick your Mac apps, get one installer.

Ninite for macOS. Tick the apps you want and get a single shareable one-liner that installs the latest version of each behind one password prompt — plus a hardened, checksum-verified MDM path for zero-touch fleet deployment.

  • Vite ESM
  • Express + Redis
  • Installomator
Flagship Enterprise / infra

ATLAS

Monitor 1,000+ Macs from one secure dashboard.

An enterprise macOS fleet-monitoring platform: real-time CPU/memory/disk/network metrics, deep network diagnostics, remote commands, and health alerts — all end-to-end encrypted with AES-256-GCM and role-based access.

  • Python 3.12
  • AES-256-GCM E2EE
  • JWT / RBAC
Native on iOS & Mac
FishTrackr app icon
iOS · iPadOS On-device ML

FishTrackr

Snap a photo, and AI logs your catch.

An angler's catch journal built on a one-tap Quick Catch flow: photograph a fish and an on-device Core ML classifier names the species — re-ranked by where you're standing. Every catch is auto-enriched with weather, water temp, pressure, and moon phase, then organized into expeditions, trips, and season stats.

  • SwiftUI + SwiftData
  • Core ML + Vision
  • WeatherKit
Privacy policy
DualLink Monitor app icon
macOS · menu bar Network diagnostics

DualLink Monitor

Wi-Fi & AirPlay link health, live in your menu bar.

A privacy-first menu-bar app that samples Wi-Fi telemetry once a second and tracks it against your AirPlay link to an Apple TV — isolating last-hop latency and flagging trouble with statistical anomaly detection before you see the stutter. Everything stays on-device; no accounts, no telemetry.

  • Swift 6 + SwiftUI
  • CoreWLAN
  • Network.framework

Who we are

Two founders, hands on the keyboard

No account managers, no handoffs. You work directly with the people building the software.

Sam Raths

Co-founder · Engineering

Full-stack engineer across JavaScript, TypeScript, and Python. Builds the studio's web and mobile products end to end — from data model to deploy.

GitHub ↗

Erik Arnold

Co-founder · Software Engineer & Data Scientist

Machine learning and HCI, full-stack. Builds the models and the interfaces around them. Lately: prediction markets, interpretable AI, and forecasting.

Get in touch

Have something you want built?

We take on a small number of projects at a time. Tell us what you're working on.