Welcome to the Daft blog
Join us as we explore innovative ways to handle multimodal datasets, optimize performance, and simplify your data workflows.
Processing 99% of U.S. Caselaw for Under $1 in the Common Pile
How Teraflop AI processed 7 million court documents and 40 million pages spanning 365 years of U.S. caselaw for under a dollar using Daft.
Agentic systems are just query engines for unstructured data
A systems engineer’s view of the new AI stack
Cutting LLM Batch Inference Time in Half: Dynamic Prefix Bucketing at Scale
A new inference backend that maximizes batch inference throughput.
Benchmarks for Multimodal AI Workloads
Spark, Ray Data, and Daft
Introducing Flotilla: Simplifying Multimodal Data Processing at Scale
Daft's new distributed engine
Exploring Daft's Local Execution
The Swordfish Engine
After the First Run
Using Daft’s observability tools to uncover performance pitfalls
Making GPUs Zoom (Part 1)
A deep dive into GPU optimizations for production-scale multimodal data processing
End-to-End Distributed PDF Processing Pipeline
OCR, Spatial Analysis & GPU Embeddings with Python
How to Build Scalable, End-to-end Batch Inference Pipelines with Daft
From prompts to parquet: making batch inference simple, fast, and scalable.
Embedding Millions of Text Documents With Qwen3
Near-100% GPU Utilization
Processing 300K Images Without OOM
A Streaming Solution
