←
Dave Zilberman1,017 posts
Dave Zilberman avatar

Dave Zilberman

@Zilberman

Anchored by Stoicism & Karma, enterprise focused VC @NorwestVP

◷ Joined August 2008▧ 79 Photos and videos
Dave Zilberman avatar
Dave Zilberman@Zilberman·Nov 15

🔥🔥🔥🔥🔥

ParseBench-2 hit 30k monthly downloads in 3 days, and is #3 on @huggingface trending benchmarks! 📈 In ParseBench-2, we tried hard to create a comprehensive taxonomy of metrics to evaluate document processing, while creating a severity-based framework to align these metrics with human judgment. Whether you’re deeply familiar with document OCR and document OCR benchmarks, or just getting started, we encourage you to check out the ParseBench-2 methodology. We’re happy to help you pick whatever subset of metrics you care about if you’re looking to process documents in production. We’ve merged dozens of PRs since the release date, including fixes/metric updates from some of our peers. We’re aggressively fixing any lingering annotation and metric issues. Report: llamaindex.ai/files/parseben… Hugging Face: huggingface.co/datasets/llam… Github: github.com/run-llama/ParseBe…

Post attachment
Jerry Liu avatar
⟳ Dave Zilberman retweeted
Jerry Liu@jerryjliu0·Nov 12

Today we’re introducing ParseBench-2 - our most ambitious document parsing benchmark to date. It is the only benchmark that provides both extreme granularity into every single type of production parsing error, while also making sure that metric scores are highly aligned with human judgment. We do this by painstakingly assigning severity scores to 330+ error types, and tuning each metric to match the severity scores. Existing benchmarks either use brittle deterministic metrics (e.g. OmniDocBench) or LLM-as-a-judge (ParseArena), which can help recognize equivalent representations but are coarse/stochastic/subject to bias by the prompt+model. ParseBench-2 also contains: ✅ 4.3k+ document pages (double from ParseBench-1). ✅ An extra dimension (forms) ✅ Much better ground-truth across all dimensions compared to ParseBench-1. This is the next evolution of ParseBench, our hugely popular benchmark that has 50k+ monthly downloads, over 150+ OCR solutions benchmarked, and has been adopted by companies like NVIDIA, Cohere, Aleph Alpha, Perplexity, and others. Alongside ParseBench-2, we are also releasing OCRia-1.5 - a new family of document parsing models that powers all our modes in LlamaParse. They represent a massive step up in capabilities from our latest release, and are state of the art in parsing complex tables, forms, charts, formatting, and assigning bounding boxes to complex layouts. Come check out our blog post, technical report, and website. Blog: llamaindex.ai/blog/introduci… Technical Report: llamaindex.ai/files/parseben… Website: parsebench.ai/ If you’re interested in trying out our OCRia-1.5 capabilities, come sign up for LlamaParse: cloud.llamaindex.ai/

2:47
Dave Zilberman avatar
Dave Zilberman@Zilberman·Nov 12

In other news… Norwest 18 is here, it’s $4B and the largest fund in @NorwestVP’s history. The new fund gives us even more opportunities to back founders building for the emerging generation of enterprise AI, data, and infrastructure.

Norwest @NorwestVP Nov 12 Article Introducing Norwest 18: $4 Billion for Every Stage of Your Journey By Jeff Crowe and Jon Kossow “Can I run something by you?” We’ve heard versions of this question countless times over the years. Sometimes it comes during a board meeting, but more often, it’s an

Dave Zilberman avatar
Dave Zilberman@Zilberman·Oct 20

We've believed for years that data loss prevention was a category ripe for reinvention. What @MilnerNitay and the ORION team have built is exactly the kind of innovation we knew was possible.

Using AI to combat AI 🥊 AI is introducing new risks to enterprise data security. It's also enabling an entirely new approach to solving them. At Black Hat 2026, ORION Security's @MilnerNitay and Norwest's @Zilberman discussed how AI is reshaping data protection. Listen in ⬇️

1:24
Dave Zilberman avatar
Dave Zilberman@Zilberman·Oct 2

What a fun ride with the @tursodatabase team! Thoroughly enjoyed partnering with @glcst, @penberg and the Turso team since 2022! And today is an exciting new chapter in the Turso journey as they join @Supabase 🚀 @NorwestVP prnewswire.com/news-releases…

Jerry Liu avatar
⟳ Dave Zilberman retweeted
Jerry Liu@jerryjliu0·Sep 20

Introducing DocJev - a lightning-fast OSS library for document classification and splitting with jev ⚡️ Give a document alongside some natural language category rules. Jev will predict the document category (classify) or the boundaries between sub-documents (split). It is 6x faster than gpt-5.6-luna with equivalent accuracy. You can choose between different OCR backends: liteparse: the fastest/most accurate free+OSS text parser out there. Perfect for digitalized documents. The speedups vs. luna above are inclusive of liteparse times. LlamaParse: the most advanced VLM-based solution for complex docs. Using this will of course add latency on document preprocessing. But it’s a good option if you want to cache the digitalized representations of documents for a variety of downstream tasks. Check it out: github.com/jerryjliu/docjev If you want to dive deeper into liteparse or LlamaParse check the links below: liteparse: github.com/run-llama/litepar… LlamaParse: cloud.llamaindex.ai/

0:24
Jerry Liu avatar
⟳ Dave Zilberman retweeted
Jerry Liu@jerryjliu0·Sep 3

Today, we're excited to launch Extract Turbo ⚡️ - the fastest VLM-powered document extraction solution. It is 3-5x faster than all other comparable OCR solutions, including our other LlamaParse tiers, at equivalent or higher accuracy.

Post attachment

Document extraction just got a new gear. ⚡️ Introducing Turbo mode for Extract, our fastest way yet to pull structured data from documents. It runs roughly 4× faster than our Cost Effective Tier at comparable accuracy, with a median latency of 3.7 seconds a page. Turbo processes pages in parallel, so latency stays nearly flat as document size grows. Available in beta today. Read the full breakdown → llamaindex.ai/blog/introduci…

Post attachment
Dave Zilberman avatar
Dave Zilberman@Zilberman·Jul 28

Verifiable Workflows > Durable Workflows If a workflow isn't verifiable, who knows if it's durable...

Durable execution is now Agentic: introducing Diagrid Catalyst 2.0. Now teams can make their existing agent workflows resilient to failure and cryptographically provable, without re-architecting how those agents are built. Agents automatically resume from the exact point of failure, and every step can be cryptographically signed and traced back to its source. It brings durable and verifiable execution to the AI frameworks developers already use: LangGraph, Microsoft Agent Framework, Google Agent Development Kit (ADK), AWS Strands, OpenAI Agents SDK, CrewAI, and more. Most of those frameworks have no built-in way to guarantee a long-running workflow recovers and finishes when something breaks. Work is lost, tokens burn, and someone has to step in. Keep the framework you build on. Add a Diagrid code package, gain durable workflows immediately. Read the release 👇 hubs.ly/Q04q_2Pb0 #AIAgents #DurableExecution #PlatformEngineering #LLMOps #DiagridCatalyst

Post attachment
Jerry Liu avatar
⟳ Dave Zilberman retweeted
Jerry Liu@jerryjliu0·Jun 2

We Parse PDFs We spent 7 figures to put this on billboards throughout SF. I thought long and hard about putting something more creative and whimsical. But then you wouldn’t know what we do. AI agents (and humans) are consuming exponentially more documents as they do real work. They need the best quality document parser to not output garbage on downstream tasks. This is what we do today as a company. If you have any PDFs (or other documents), we parse them :) If you’re around SF in June for one of the following events, come stop by our booths: ✅ Snowflake Summit (this week, Booth 1123) ✅ Databricks Data+AI Summit (June 15-18, Booth 137) ✅ AI Engineer World Fair(June 29-July 2, Booth L-G47) You can find us by the same sign we put on our billboards! We Parse PDFs @llama_index

2:26
Dave Zilberman avatar
Dave Zilberman@Zilberman·May 28

Invest and sign a definitive agreement in the same quarter? Well, this is a new one! Congrats to @pratyus and the @natomalabs team on entering into a definitive agreement with @Snowflake. The demand for agentic AI infrastructure is moving fast. natoma.ai/blog/natoma-snowfl…

Jerry Liu avatar
⟳ Dave Zilberman retweeted
Jerry Liu@jerryjliu0·May 23

If you're stopping by the SF Caltrain station over Memorial Day weekend, you might catch a glimpse of our digital ads 📺 We parse (PDFs) (50+ other document types)

Post attachment
Glauber Costa avatar
⟳ Dave Zilberman retweeted
Glauber Costa@glcst·May 4

Turso now includes unlimited active databases in every plan. We already had unlimited databases, but we would charge you based on how many of them were active. That is now gone. You want a database, you get a database.

0:39
Jerry Liu avatar
⟳ Dave Zilberman retweeted
Jerry Liu@jerryjliu0·Apr 14

This is why we released liteparse :) Free, open-source, designed for agents. Natively supports OCR / screenshotting for deeper visual understanding in a document when needed.

I just tried it this morning on the 245-page Mythos pdf and it failed badly and the outputs were all mangled. Converting pdfs is really hard, I think it has to probably be a Skill not a program, for a SOTA LLM for it to work properly.