Declarative Context Engineering for Agents
Transform unstructured and structured data into insights using familiar DataFrame operations enhanced with semantic intelligence. Built for AI and agentic applications.
Quick Start
Get up and running with Fenic in minutes.Installation
Install Fenic and set up your environment
Quickstart Guide
Build your first semantic data pipeline
Key Features
Fenic brings the reliability of traditional data pipelines to AI workloads.Semantic Operators
LLM-powered transformations: extract, classify, map, join, and reduce data with natural language
PySpark-Inspired API
Familiar DataFrame operations with lazy evaluation and query optimization
Unstructured Data Support
Native support for markdown, transcripts, JSON, and PDF processing
Multi-Provider LLMs
OpenAI, Anthropic, Google, Cohere, and OpenRouter with automatic batching
Batch Inference
Automatic batch optimization, rate limiting, and cost tracking
Production Ready
Built-in retry logic, error handling, and comprehensive observability
Core Concepts
Learn the fundamental concepts behind Fenic.Sessions
Manage execution context and configuration
DataFrames
Work with structured and unstructured data
Semantic Operators
Apply LLM-powered transformations
Examples
Explore real-world use cases and patterns.Hello World
Introduction to semantic extraction and classification
Document Extraction
Extract structured data from PDFs and documents
Semantic Joins
Match data across tables by meaning, not just values
Meeting Transcripts
Process and analyze meeting transcripts with speaker awareness
Community & Support
GitHub
Star us on GitHub and contribute to the project
Discord Community
Join our Discord community for help and discussions