DashBlox | Devpost

Target Sponsor Tracks

Track / Company	How DashBlox aligns
ElevenLabs - Best Use of ElevenLabs	Core production feature. We generate all game SFX through ElevenLabs and also support TTS + voice cloning workflows.
Google Gemini - Best Use of Gemini API	We use Gemini (via OpenRouter) to enhance model/audio prompts into production-ready generations before hitting 3D/audio providers.
MongoDB Atlas - Best Use of MongoDB Atlas	DashBlox persists project metadata, generated asset records, board state, and workflow history using MongoDB models.
Vultr - Best Use of Vultr	Deployment target for hosting the dashboard API/frontend as a production-ready cloud service.
Reach Capital - Best Use of AI powered by Reach Capital	DashBlox applies AI to upskill creators by reducing technical overhead in asset creation, planning, and iteration workflows.

Inspiration

Game development is slow because teams constantly jump between disconnected tools for 3D assets, audio, planning, docs, and implementation. We built DashBlox to eliminate that fragmentation and prove that a single AI-native pipeline can help teams ship complete, polished games faster.

What it does

DashBlox is an AI game production system designed as a one-stop shop for game development. It lets teams generate custom meshes, create game-ready SFX and voices, run production in a built-in Trello-style board, and manage living project docs in one place.

We validated the platform by building a full wave-based zombie game where:

all game assets were custom meshes created through our mesh workflow
all planning was done in our dashboard Board
all game SFX were generated with ElevenLabs
voice generation and voice cloning are supported for dialogue and character content

How we built it

We built a fullstack web platform (Node.js + Express + vanilla frontend + MongoDB models) with five core modules:

Models: prompt enhancement (Gemini via OpenRouter), multi-provider generation (Meshy/Tripo/Rodin/Replicate), preview, conversion, and export workflows
Converter: format conversion pipeline for game-engine-ready assets
Audio: ElevenLabs SFX generation, TTS, voice library, and voice cloning
Docs: editable project knowledge base for PLAN.md and PROGRESS.md files
Board: kanban workflow (our internal Trello equivalent) for sprint execution

Then we connected the pipeline directly to our build loop: generate assets and sounds, prioritize on the board, integrate in-game, playtest, and iterate.

Challenges we ran into

Normalizing very different AI provider APIs into one reliable generation UX
Handling async generation and polling while keeping the dashboard responsive

Accomplishments that we're proud of

Shipped both the platform and a complete polished game using it
Created a true end-to-end workflow from prompt to in-game asset/SFX integration

What's next for DashBlox

One-click publish pipeline from generated asset to engine/package-ready delivery
Shared asset history with versioning and rollback
Team collaboration upgrades (activity timeline, live board sync enhancements)