Welcome to DODIL
DODIL is building a comprehensive AI cloud — an end‑to‑end platform that gives engineers a single, consistent stack across the full lifecycle of AI-native software.
This is the documentation hub.
Services
Each product has its own guides, API reference, and workflows — open the one you need.
DataK³
K³ — one bucket, every dimension
A multi-dimensional data plane. Upload once, then query the same data as objects, rows, embeddings, or a graph — over S3, Postgres, gRPC, Pinecone/Qdrant, Bolt, and HTTP.
- Objects, tables, vector, and graph in one bucket
- Wire adapters: S3, Postgres, Bolt, Pinecone, Qdrant
- Automated ingest pipelines
Ignite
Serverless apps & inference
Deploy stateless functions and containers. Bring code, an image, or a git repo — the platform builds, deploys, and scales it. Includes the model catalog for LLMs, embeddings, and rerankers.
- Functions and containers, zero infra
- OpenAI/Cohere-compatible model APIs
- Build from git, upload, or prebuilt images
Scriptum
Workflow orchestration
Typed, durable workflows and agents. Mix deterministic steps with agent reasoning — checked before it runs.
- A typed workflow language
- Durable threads with human-in-the-loop
- Template catalog for common flows
Git
Source control
Git-powered repositories with pull requests, diffs, and CI checks. Clone and push over smart-HTTP or SSH; a REST control plane drives everything else.
- Smart-HTTP and SSH git wire
- Pull requests, reviews, CI checks
- SSH keys and org API keys
Registry
Container registry
Store and distribute container images with vulnerability scanning, tags, and per-repo visibility. Standard Docker Registry HTTP API — docker login, push, pull.
- Docker Registry v2 API compatible
- CVE scanning at push time
- Public and private repositories
AppID
Identity for your apps
Authentication and identity for the applications you build — sign-in, users, and tokens, without running your own identity stack.
- OIDC-standard auth flows
- Hosted, brandable sign-in UI
- Users and tokens per app
Learn more
- Platform Design — how the platform is built: RPC-first APIs exposed over HTTP, and agent-friendly discovery via reflection.
- Use Cases — what teams build on DODIL, with the “why Dodil” for each. For step-by-step walkthroughs, see the blog .