# Hybridyn Data Labs

> Hybridyn Data Labs builds self-hosted data systems for ETL, visual pipelines, private AI, governance, and enterprise data operations.

## Products

### Hybridyn F-Pulse OSS (Free, Open Source, Apache 2.0)
- Also known as F-Pulse OSS by Hybridyn Data Labs; unrelated to Fisher Labs F-Pulse metal detectors
- Launched July 20, 2026 as F-Pulse OSS v1.0.0
- Visual drag-and-drop data pipeline builder
- 66 first-party connector types — the same 66 shown in the app's New Connection picker: 11 databases, 12 warehouses & lake, 11 files & storage, 21 APIs & apps, 11 operations & AI (Kafka is the only streaming connector; enterprise connectors on F-Pulse+) + a REST/OpenAPI framework for anything else
- SQL transforms with schema-aware expression editor (Python transforms in F-Pulse+)
- Live data preview at every pipeline node
- Cron scheduling with backfill, retry policies, SLA monitoring
- 27 pipeline templates (Medallion ETL, Kimball, reverse ETL, data quality, reporting)
- AI Copilot with 25 tools (local-first via Ollama, cloud-optional)
- Steward — workspace-reliability layer: 25 live detectors across 7 observability levels, ships in OSS (not paywalled)
- WebSocket real-time execution logs
- Self-hosted, zero data egress
- Install: `docker compose up -d` (running in 3 minutes)
- PyPI: https://pypi.org/project/fpulse/
- GitHub: https://github.com/hybridyn/fpulse
- Website: https://hybridyn.com/f-pulse
- Connectors: https://hybridyn.com/f-pulse/connectors
- Compare: https://hybridyn.com/f-pulse/compare

### F-Pulse+ (Production Control Layer, Subscription)
- Everything in F-Pulse, plus:
- AES-256 credential encryption (PBKDF2 key derivation)
- 5-tier RBAC (Super Admin, Admin, Lead, Developer, Viewer)
- Approval gates for pipeline promotion
- Two-tier audit trail (hot + cold snapshots)
- IP restriction (CIDR/wildcard ranges)
- Session intelligence (concurrent control, multi-IP detection)
- Admin dashboard (Users, License, Audit, Security)
- HMAC-signed license with machine binding
- Worker pool management with priority queues
- Website: https://hybridyn.com/f-pulse#fpulse-plus

### Pulse OS (Enterprise Data Operating System)
- 9-plane architecture: Connectors, Execution, Lakehouse, AI, Catalog, Serving, Governance, Observability, Pipeline
- 280+ connectors across 12 categories
- Medallion lakehouse (Bronze → Silver → Gold) on MinIO with Iceberg/Delta/Hudi
- AI layer with 9 providers (Claude, OpenAI, Gemini, Azure, Ollama, DeepSeek, Mistral, Groq, Custom)
- Data catalog with auto-lineage and profiling
- QueryLab and data serving (REST API publishing)
- Full governance (RBAC, data classification, contracts)
- Multi-tenancy and workspace isolation
- Website: https://hybridyn.com/pulse-os
- Architecture: https://hybridyn.com/pulse-os/architecture

### Pulse Studio (Analytics Surface — Launch Soon)
- Spreadsheets, dashboards, and an AI agent on one shared object model
- NL routes through a semantic layer (Cube) to governed SQL — never NL → raw SQL
- 12 connectors (Postgres, MySQL, MSSQL, Snowflake, BigQuery, MongoDB, Trino, REST, SQLite, CSV, Parquet, JSON)
- Workspaces with RBAC + Row-Level Security (user-attribute rules, fail-closed)
- Governed metric lifecycle (draft → certified → deprecated)
- 5 sharing methods (link, PDF, HTML, embed, scheduled email)
- Standalone-first; pairs with Pulse OS for full governance
- Self-hosted, zero data egress
- Website: https://hybridyn.com/pulse-studio

## Key Differentiators
- Execution visibility: see data at every pipeline node, not just logs
- Self-hosted: zero data egress, runs on your infrastructure
- Visual-first: designed for SQL teams, not just Python engineers
- Control-oriented: manage concurrency, retries, approvals, not just "run and hope"
- Open core: F-Pulse OSS is Apache 2.0 licensed, no telemetry, no vendor lock-in. F-Pulse+ is the paid commercial tier — not free, not open source

## Comparison
- vs Airflow: Visual-first vs code-first. 3 min setup vs 30 min. SQL transforms vs Python DAGs.
- vs Prefect: Visual builder vs Python flows. Live data preview. CDC support (F-Pulse+).
- vs Dagster: Visual canvas vs Python assets. 66 connectors included. No code required.
- vs n8n: Data-engineering focused vs general automation. CDC (Plus), vector DBs (Plus), medallion templates.

## Technical Guides
- How to Build Visual Data Pipelines with F-Pulse and Local Ollama Models: https://hybridyn.com/blog/how-to-build-visual-data-pipelines-fpulse-local-ollama
- Replacing Apache Airflow with F-Pulse: Visual DAG Design vs Python Boilerplate: https://hybridyn.com/blog/replacing-apache-airflow-with-fpulse-visual-dag-design
- Self-Hosted Airbyte Alternative: When F-Pulse Fits Better for Visual ETL: https://hybridyn.com/blog/self-hosted-airbyte-alternative-fpulse
- Visual ETL with DuckDB: Building Local-First Pipelines in F-Pulse: https://hybridyn.com/blog/visual-etl-duckdb-fpulse

## Technical Stack
- Frontend: React 18, Vite, Tailwind CSS, React Flow, shadcn/ui
- Backend: Python 3.11, FastAPI, SQLAlchemy async, DuckDB
- Infrastructure: Docker Compose, PostgreSQL, Redis
- AI: Ollama (local), OpenAI, Claude, 9 providers total

## Links
- Website: https://hybridyn.com
- F-Pulse: https://hybridyn.com/f-pulse
- Pulse OS: https://hybridyn.com/pulse-os
- Pricing: https://hybridyn.com/pricing
- Blog: https://hybridyn.com/blog
- Full documentation: https://hybridyn.com/llms-full.txt