What is Radhflow?

Radhflow is a compiled agent platform. You describe a data pipeline. AI generates typed, deterministic code — SQL transforms, CLI scripts, API connectors. That code runs without any LLM in the loop. Zero token cost at runtime. Same input, same output, every time.

Pipelines live as YAML + scripts in a Git repository. You own every file.

Architecture

 You ──describe──▶ ┌────────────┐    ┌────────────────┐
                   │  Conductor │───▶│  Code Agents    │
                   │  (AI)      │    │  (AI, one-time) │
                   └────────────┘    └───────┬────────┘
                                             │ generates
                                             ▼
                   ┌─────────────────────────────────────┐
                   │  Pipeline (YAML + scripts in Git)   │
                   │                                     │
                   │  ┌──────┐   ┌──────┐   ┌──────┐    │
                   │  │ Node │──▶│ Node │──▶│ Node │    │
                   │  └──────┘   └──────┘   └──────┘    │
                   └─────────────────────────────────────┘
                                             │ executes
                                             ▼
                   ┌─────────────────────────────────────┐
                   │  Deterministic Runtime               │
                   │  DuckDB · Nix CLI · HTTP · Browser   │
                   │  No LLM. No tokens. No surprises.    │
                   └─────────────────────────────────────┘

AI touches your pipeline once — at creation time. After that, it is plain code.

Core properties

• Deterministic execution. No LLM at runtime. Pipelines produce identical output for identical input.
• Four data types. Value (scalar), Record (single object), Table (rows), Stream (unbounded rows). Nothing else needed.
• NDJSON interchange. Tables are .ndjson files with .schema.json companions. Human-readable, diffable, universal.
• DuckDB transforms. SQL is the default language for filtering, joining, aggregating. No custom DSL.
• Nix-managed CLI nodes. Shell tools declared in Nix. Reproducible across machines.
• Git-native. Branch a pipeline. Diff it. Review it in a PR. Merge it.
• Local-first. Runs on your laptop. Cloud deployment optional. EU infrastructure when you need it.

Who it’s for

Role	Use case
Marketing coordinator	Lead scoring, campaign reporting, CRM sync — no engineering tickets.
Data engineer	Typed, reproducible ETL pipelines versioned in Git.
VFX TD	Asset processing pipelines with CLI tools managed by Nix.
Freelancer	Client data workflows that run locally without cloud subscriptions.
Small business owner	Automate invoicing, inventory, reporting — own the pipeline forever.

What Radhflow is NOT

Not a data warehouse. Radhflow moves and transforms data. It does not store petabytes. Use it alongside Postgres, BigQuery, or a lakehouse.

Not an agent framework. There is no autonomous agent loop at runtime. AI generates code once. That code is static and deterministic.

Not a no-code tool. Radhflow generates real code. You can read, edit, and extend every file. The visual canvas is a view into YAML, not a replacement for it.

Not US-cloud-dependent. Local-first by default. Cloud deployment runs on EU infrastructure. Your data stays where you put it.