ragbits and Building-Natural-Language-and-LLM-Pipelines
These are complementary tools: ragbits provides reusable building blocks and abstractions for RAG/agentic systems, while the Packt book offers practical implementation patterns and recipes demonstrating how to construct such systems using specific frameworks (Haystack, RAGAS, LangGraph).
About ragbits
deepsense-ai/ragbits
Building blocks for rapid development of GenAI applications
Provides modular Python packages for LLM integration (100+ models via LiteLLM), RAG pipelines with 20+ document formats, and multi-agent coordination using the A2A protocol and Model Context Protocol. Features type-safe prompt execution with Python generics, support for Qdrant/PgVector and other vector stores, Ray-based distributed document ingestion, and OpenTelemetry observability—installable as granular components (core, agents, document-search, evaluate, guardrails, chat, CLI) rather than monolithic framework.
About Building-Natural-Language-and-LLM-Pipelines
PacktPublishing/Building-Natural-Language-and-LLM-Pipelines
Building RAG and Agentic Applications with Haystack 2.0, RAGAS and LangGraph 1.0 published by Packt
Covers deterministic pipeline design with strict tool contracts, context engineering for agent reliability, and production deployment patterns including microservices via FastAPI/Hayhooks and multi-agent orchestration with LangGraph's supervisor-worker patterns. Integrates evaluation frameworks (RAGAS, Weights & Biases) for cost and quality tracking, plus practical NLP tasks like NER and sentiment analysis as agentic tools within observable, fault-tolerant workflows.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work