← Back to Portfolio

Slingshot

Deep Space Operations Simulator

The Vision

A space operations game where you only know what your sensors tell you.

Not a point-and-click strategy game with perfect information. An epistemic simulation where information is earned, uncertain, and valuable.

You scan an asteroid. Your radar says it's 200 meters wide. Your optical says 180. Your spectrometer detects iron, but with 60% confidence. Do you believe it? Do you mine it? Do you sell the data before it goes stale?

This is Slingshot.

What We Built

  • Full orbital mechanics simulation with GPU-accelerated trajectory computation
  • Epistemic sensor system -- 8 instrument types (radar, optical, IR, spectrometer, LIDAR, magnetometer, star tracker, particle detector), each with unique displays and failure modes
  • Knowledge economy -- data has real value, decays over time, and must be sold before it goes stale
  • Research system with 98 technology lines across 10 categories, diminishing returns with breakthrough moments
  • Persistent asteroid belt with millions of objects rendered via GPU compute shaders
  • Ship design and construction pipeline -- design, build, launch, operate
  • NPC companies as full player-equivalent actors with long-term strategies
  • Multi-agent AI development pipeline for rapid feature building

Design Philosophy

You Only Know What Your Sensors Tell You

No omniscient view. No perfect information. Every piece of data is measured, uncertain, and valuable. Scan, cross-reference, verify.

Grounded Science Fiction

Realistic physics, plausible technology progression. No faster-than-light drives, no magic shields. Just Newtonian mechanics and clever engineering.

Multiplayer from the Core

Not bolted on after the fact. Every system designed for player-vs-player interaction, competition, and cooperation from day one.

Data-Driven Design

Game balance lives in data files, not code. Ship stats, tech trees, sensor parameters -- all tunable without recompilation.

Iterative Engineering

Ship designs should evolve as research progresses. Early craft are clunky, inefficient. Late-game vessels are refined, purpose-built machines.

Technical Highlights

GPU Compute Shaders

Real-time orbital calculations and asteroid field rendering offloaded to the GPU for smooth performance.

Multi-Agent AI Pipeline

Parallel feature development using AI agents -- design, prototype, test, iterate at high velocity.

Centralized UI System

Consistent visual language across all interfaces. Every panel, every readout, same design principles.

Signal-Based Architecture

Clean system decoupling via event-driven signals. Components communicate without tight coupling.