Portfolio / 2026 · Ansbach, DE

Moritz Bäuerlein

I build games that swing and study networks that learn. This little unit is my guide: scroll and it will show you around.

(01) Selected work · Case 01 · Game

FlyHigh key art: stone letters on floating islands with a hovering robot boss

FlyHigh

A high-speed multiplayer grappling platformer. Swing across floating islands, dodge bombs, lava and hunters, and race your friends to the top of the sky.

Engine
Unreal Engine (C++ / Blueprints)
Players
Up to 4, PvP race or co-op
Backend
Steam Advanced Sessions, own leaderboard
Robot Run · play me 0000 / best 0000
A

A vertical multiplayer playground

Players grapple upward until they reach the pinnacle of the sky, across a massive obstacle course of floating islands, each with its own hazards and bounce properties.

B

Authoritative momentum

To make a physics-based hook feel instant over the internet, the swing uses client-side prediction and server reconciliation: the client simulates the arc with zero input lag while the server stays the authority.

C

Dynamic threats, competitive backend

AI enemies, bombs, lava and moving obstacles keep players swinging. Localized navigation invokers build pathing bubbles only around active players. Runs end on a global leaderboard that I built and host myself.

A

Practical applications

It started with Python scripts that used AI for tasks plain code struggled with. Highlights: Clawdbot on a Raspberry Pi talking to me through Telegram, and a voice assistant app wired to a local model, image and search APIs.

B

Theory and mathematics

Backpropagation and gradient descent tore back the "veil of magic". I built text recognition, KNN classification and neural networks from scratch with plain Python lists before moving to PyTorch.

C

Foundations from school

Computer architecture from the single bit up: logic gates, carry adders, flip-flops and automata theory. Then tokens and embeddings, convolution kernels, and small models in scikit-learn.

Case 02 · Research & study

Illustration of a circuit-board brain surrounded by logic gates

Learning machines

From automation scripts on a Raspberry Pi to neural networks written with plain Python lists. Now studying Applied AI at university.

Study
B.Eng. Applied AI, Hochschule Ansbach
Stack
Python, PyTorch, scikit-learn
Focus
Backprop, KNN, assistants

(02) About

About me

I'm a curious developer who builds games and studies Machine Learning. You can create almost anything with numbers and logic, and that's why I love it.

Programming is how I bring ideas to life and solve problems creatively. In my spare time I work on personal projects, mostly games or experiments with AI. I'm always open to a good conversation.

I got into programming through small Python scripts and never really stopped. FlyHigh, my multiplayer grappling game, taught me Blueprints, multiplayer replication, procedural worlds and what it takes to actually finish and release a project. At Hochschule Ansbach I'm now learning the maths behind the models I used to just play around with.

Away from the screen you'll find me behind a drum kit, building Lego Technic or in the gym.

Moritz sitting relaxed, one hand at his temple Moritz standing and typing on a keyboard in the air
  • NowB.Eng. Applied AI, Hochschule Ansbach
  • BuiltFlyHigh, a multiplayer grappling game
  • ToolsUnreal, Python, PyTorch, Fusion 360, DaVinci Resolve
  • Off-screenDrums, Lego Technic, strength training

(03) Contact

Your details are only used to answer your message. Privacy