Minecraft Bedrock Developer
I build production-scale Minecraft Bedrock systems, from Script API gameplay architecture and custom UI to server technology and competitive multiplayer.
Bedrock systems I have designed, shipped and maintained, with the technical decisions behind them.
A recreation of Java Edition's 1.9 combat model on Bedrock. Timed hits, attack cooldown, sweeping edge, Java knockback and shield behaviour, shipped as a stable addon with over a million downloads.
A competitive Bedrock PvP network on a custom multi-server architecture, with proxy-based server transfers, matchmaking, ranked play and database-backed player profiles.
Bedrock does not give you most of what these projects needed. Four constraints and what I built around them.
Constraint. Bedrock has no attack cooldown. Its combat is click driven, while Java's 1.9 model makes damage a function of how far an attack has recharged.
Approach. Rebuilt the charge state machine in the Script API, derived its recharge rate per item, and reconciled it with the engine's own damage event so vanilla and scripted damage never double apply.
Constraint. Bedrock exposes no custom HUD interface. An attack indicator and a saturation bar are standard Java UI with nothing to hook into.
Approach. Drew both through the resource pack UI layer, driven by the same script state that governs combat, so the display never drifts from the mechanic behind it.
Constraint. The easy route to Java combat on Bedrock runs through experimental toggles or a custom player.json. Both break other content.
Approach. Ruled both out and built the whole feature set from the stable scripting and resource pack surface. A smaller toolset, but the reason the addon can sit alongside other creators' work.
Constraint. Bedrock has no standard proxy layer, so the multi-server pattern that makes competitive networks routine on Java does not exist.
Approach. Building a custom proxy that owns the client connection and points it at another instance, with match lifecycle and player identity moved out of server memory into MariaDB.
I am a Minecraft Bedrock developer. Most of my experience comes from building and running my own projects rather than from a job, which has meant owning the whole lifecycle: design, implementation, release, bug reports from real users, and keeping things working as Minecraft changes underneath them.
Blurry's Java PvP has been my main body of work since 2023. Rebuilding Java's combat model on a platform with no attack cooldown, no sweep attack, no saturation stat and no HUD API meant writing each of those systems myself, without experimental toggles or a custom player.json, so the addon stays compatible with other people's content.
Paragon is where that work moves past addon scripting into server infrastructure: a multi-server Bedrock network with a custom proxy layer, Endstone plugins in Python, and player data in MariaDB. It is still in development, and the project page marks what is built and what is not.
I want to do Marketplace work because the constraints there are the ones I already build under: production quality, compatibility, performance, and systems someone else has to maintain.
What I build with, and where each of these has actually been used.
The core of my work. Gameplay systems built on the Script API and shipped as production addons.
Bedrock server technology beyond the addon layer: multi-server networks and the proxy work they require.
Languages I actually ship in.
Backend systems for persistent player data and multi-server infrastructure, beyond the pack runtime.
I am open to Bedrock development roles, Marketplace studio work, contract development and technical collaboration.