Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

Why aren't more people talking about how difficult it is to turn documents into structured data? This is literally a problem that every single company I talk to is trying to solve. They have a buttload of documents with forms and tables, and they want to turn them into... show more

Santiago

452,241 subscribers

46,405 views • 1 year ago •via X (Twitter)

Science & Technology Education

Anya Rossi• Live Now

Private livecam show

10 Comments

Santiago1 year ago

There are two awesome things here: First, you can use @tensorlake's Document Ingestion API to process all your files. They tell me they have a 98% - 99% accuracy processing insurance and bank documents, which are usually a nightmare. Second (and this is what I love the most), you can turn those tables and forms into structured data. You start with the image of a form and end with a JSON file containing the information you wanted to extract. You should definitely try this out: 1. Go to 2. Sign up 3. Try your documents in the playground No credit card required for any of this, and you have plenty of credits to try.

ibexdream1 year ago

Companies sit on goldmines of PDFs, scans, and form, but can’t extract value without serious effort. Tools like @tensorlake are game-changers for operational intelligence.

OJ1 year ago

My guesstimation is because most people in/around AI seem to have very little experience, in general, working at/with (big) companies. Therefore they know very little about the biz reality and the main challenges/how things actually work.

Santiago1 year ago

Yeah, this is accurate.

Santiago1 year ago

Not even close. I'm curious, why do you think this is solved by calling a model?

Piyush1 year ago

I work with clients on large volume of documents in high stake fields (healthcare, real estate, finance, etc) and where AWS textract wins is explicit provision of confidence scores which helps to ensure only high-reliability data is used in downstream processes. Will check out tensorlake too for low-stake documents if the pricing makes sense.

λthugg-huh?1 year ago

do you do much besides shilling stuff? seriously all i see is you recommending a tool thats gonna change the whole game and obsolete the tool you shilled last week

Santiago1 year ago

I'd suggest you stop following me if my posts bother you too much.

Neel Das1 year ago

This problem is more or less solved now with a simple API call to any of the reasoning models

Michael | Æ1 year ago

So true. This is such a huge challenge for so many teams. Thanks for sharing a solution that works.

Related Videos

Every company I talk to is literally trying to solve this problem: How to automatically extract structured data from unstructured documents. For example, they want to process driver's licenses, no matter the file type or format. I recorded the solution for this:

Every company I talk to is literally trying to solve this problem: How to automatically extract structured data from unstructured documents. For example, they want to process driver's licenses, no matter the file type or format. I recorded the solution for this:

Santiago

193,603 views • 11 months ago

GPT-4o is bad at processing PDF documents. Whoever tells you otherwise is not living in the real world. In 2024, people fill out forms using pen and paper. Try to answer questions from those forms using modern models, and you'll be disappointed. I recorded a video to show you how to fix this. The answer is simple: stop letting the model see your PDF document. Instead, preprocess it and stick to showing the model text. Watch the video. You'll go from "THIS IS CRAP" to "EXCELLENT" in no time. I'm using Unstract to turn the documents into text while keeping the original format (this is crucial!) You can use them to process up to 100 pages for free. They collaborated with me on this post. You can find the code I wrote right here:

GPT-4o is bad at processing PDF documents. Whoever tells you otherwise is not living in the real world. In 2024, people fill out forms using pen and paper. Try to answer questions from those forms using modern models, and you'll be disappointed. I recorded a video to show you how to fix this. The answer is simple: stop letting the model see your PDF document. Instead, preprocess it and stick to showing the model text. Watch the video. You'll go from "THIS IS CRAP" to "EXCELLENT" in no time. I'm using Unstract to turn the documents into text while keeping the original format (this is crucial!) You can use them to process up to 100 pages for free. They collaborated with me on this post. You can find the code I wrote right here:

Santiago

196,836 views • 1 year ago

Mikel Arteta. Wow. ❤️‍🔥 🗣️ “When you have a difficult period, I think the best thing you can do, instead of talking a lot, is observe. Look around you, see how people react. How they talk, how they look at you, how they judge you. Do they look at themselves, do they criticise others?” “Look around and you will learn the environment and the people that you have around you. I couldn’t be prouder to work at a club with people that ask, ‘what else can I do to help?’.” 😤 🎥 Hayters TV

Mikel Arteta. Wow. ❤️‍🔥 🗣️ “When you have a difficult period, I think the best thing you can do, instead of talking a lot, is observe. Look around you, see how people react. How they talk, how they look at you, how they judge you. Do they look at themselves, do they criticise others?” “Look around and you will learn the environment and the people that you have around you. I couldn’t be prouder to work at a club with people that ask, ‘what else can I do to help?’.” 😤 🎥 Hayters TV

DailyAFC

73,376 views • 3 months ago

"The whole idea behind WikiLeaks is to take the same technologies that allow the NSA and Google to turn you into a source of data, and turn it against them to make you more opaque and them more transparent." - Yanis Varoufakis

"The whole idea behind WikiLeaks is to take the same technologies that allow the NSA and Google to turn you into a source of data, and turn it against them to make you more opaque and them more transparent." - Yanis Varoufakis

WikiLeaks

214,888 views • 2 years ago

🇺🇸 NVIDIA CEO: PALANTIR TURNS DATA AND HUMAN JUDGMENT INTO INSIGHT “This is Palantir ontology. They take information, they take data, they take human judgment, and they turn it into business insight. We work with Palantir to accelerate everything they do — to process data at a much larger scale and speed, whether it’s structured data from the past or the real-time data shaping the future.” Source: Yahoo Finance

🇺🇸 NVIDIA CEO: PALANTIR TURNS DATA AND HUMAN JUDGMENT INTO INSIGHT “This is Palantir ontology. They take information, they take data, they take human judgment, and they turn it into business insight. We work with Palantir to accelerate everything they do — to process data at a much larger scale and speed, whether it’s structured data from the past or the real-time data shaping the future.” Source: Yahoo Finance

Mario Nawfal

192,979 views • 8 months ago

Mikel Arteta’s beautiful praise to his staff and players. “That's when you have a difficult period. I think the best thing that you can do instead of talk a lot is observe. Look around you and see how people react. How they talk, how they look at you, how they judge you, what they do. Do they look at themselves, do they start to criticise other people? “Look around and you're going to learn the environment and the people that you have around you. “I cannot be prouder to work in a club with people that the only thing they could do is ask, what else can I do to help? “When you have people like this, I don't know if it's gonna take another week or two, but something good it will happen at the end because we deserve it.”

Mikel Arteta’s beautiful praise to his staff and players. “That's when you have a difficult period. I think the best thing that you can do instead of talk a lot is observe. Look around you and see how people react. How they talk, how they look at you, how they judge you, what they do. Do they look at themselves, do they start to criticise other people? “Look around and you're going to learn the environment and the people that you have around you. “I cannot be prouder to work in a club with people that the only thing they could do is ask, what else can I do to help? “When you have people like this, I don't know if it's gonna take another week or two, but something good it will happen at the end because we deserve it.”

Connor Humm

48,588 views • 3 months ago

Everyone is building RAG applications, but nobody is talking about the data these systems use. You are delusional if you think clients will have their data sitting in a folder waiting for you to process it. Data is everywhere: in Google Drive, Dropbox, S3, Gmail, Slack, you name it. And, of course, no sane developer wants to build connections to every one of these systems. This would be suicide. I'm working with Ragie, and they released Ragie Connect to solve this problem. First, their RAG system is top-notch (they have published how they do on several RAG benchmarks), and with Connect, they made it very simple to integrate client data without having to write any code. (Well, in reality, you still have to write a few lines, but it's minimal.) Instead of developing one-off integrations for Drive, Dropbox, etc, you can use Connect to integrate with all of them and let Ragie handle authentication, authorization, and automatic data synchronization. This is a huge time saver!

Everyone is building RAG applications, but nobody is talking about the data these systems use. You are delusional if you think clients will have their data sitting in a folder waiting for you to process it. Data is everywhere: in Google Drive, Dropbox, S3, Gmail, Slack, you name it. And, of course, no sane developer wants to build connections to every one of these systems. This would be suicide. I'm working with Ragie, and they released Ragie Connect to solve this problem. First, their RAG system is top-notch (they have published how they do on several RAG benchmarks), and with Connect, they made it very simple to integrate client data without having to write any code. (Well, in reality, you still have to write a few lines, but it's minimal.) Instead of developing one-off integrations for Drive, Dropbox, etc, you can use Connect to integrate with all of them and let Ragie handle authentication, authorization, and automatic data synchronization. This is a huge time saver!

Santiago

72,976 views • 1 year ago

Here is how you can install an open-source, enterprise-grade RAG system on your server (with the best document understanding I've seen.) First, something obvious to anyone trying to sell RAG in the market: You are crazy if you think companies will let their data travel to a hosted model. No one wants to send their data anywhere (those who do haven't found an alternative.) Every single company would rather have an air-gapped system with no internet access. GroundX is an open-source RAG system that you can run on your servers (or any cloud provider, as long as you have access to GPUs) and works without a network. (If the military wants to do RAG, this is precisely what they will be looking for.) I installed GroundX on my AWS account and recorded a video to show you how to use it. There are two services you can use: 1. Ingest: This service uses a pretrained vision model to ingest and understand your knowledge base. 2. Search: This service combines text and vector search with a fine-tuned re-ranker model to retrieve information from your knowledge base. A quick note about the Ingest service: 99% of people think they need better "retrieval" mechanisms. I think they need better "ingestion." That's where this service comes in! Ingest "understands" your documents in a way I haven't seen before. After you try it, you'll realize why showing your LLM your raw documents is a bad idea. In the video, I use a free tool called X-Ray to test a document and understand how the Ingest service breaks it down. You can access this tool by signing up for a free GroundX cloud account and uploading your documents. You'll see a bit more about this in the video.

Here is how you can install an open-source, enterprise-grade RAG system on your server (with the best document understanding I've seen.) First, something obvious to anyone trying to sell RAG in the market: You are crazy if you think companies will let their data travel to a hosted model. No one wants to send their data anywhere (those who do haven't found an alternative.) Every single company would rather have an air-gapped system with no internet access. GroundX is an open-source RAG system that you can run on your servers (or any cloud provider, as long as you have access to GPUs) and works without a network. (If the military wants to do RAG, this is precisely what they will be looking for.) I installed GroundX on my AWS account and recorded a video to show you how to use it. There are two services you can use: 1. Ingest: This service uses a pretrained vision model to ingest and understand your knowledge base. 2. Search: This service combines text and vector search with a fine-tuned re-ranker model to retrieve information from your knowledge base. A quick note about the Ingest service: 99% of people think they need better "retrieval" mechanisms. I think they need better "ingestion." That's where this service comes in! Ingest "understands" your documents in a way I haven't seen before. After you try it, you'll realize why showing your LLM your raw documents is a bad idea. In the video, I use a free tool called X-Ray to test a document and understand how the Ingest service breaks it down. You can access this tool by signing up for a free GroundX cloud account and uploading your documents. You'll see a bit more about this in the video.

Santiago

89,624 views • 1 year ago

Keith Rabois on the “one person, one problem" framework he learned from Peter Thiel "Peter Thiel used to insist at PayPal that every single person could only do exactly one thing. And we all rebelled. You feel like it's insulting to be asked to do just one thing. But Peter would enforce this pretty strictly. He'd basically say: 'I will not talk to you about anything else except for this one thing that I've assigned to you. I don't want to hear about how great you're doing in this other area. Just focus until you conquer this one problem.'... The insight behind this is that most people will solve problems that they understand how to solve. Roughly speaking, they will solve B+ problems instead of A+ problems. A+ problems are high-impact problems for your company but they're difficult--you don't wake up in the morning with a solution to them, so you tend to procrastinate... If you have a company that's always solving B+ problems, you'll never create the breakthrough idea because no one is spending 100% of their time banging their head against the wall every day until they solve it" Video source: Y Combinator (2014)

Keith Rabois on the “one person, one problem" framework he learned from Peter Thiel "Peter Thiel used to insist at PayPal that every single person could only do exactly one thing. And we all rebelled. You feel like it's insulting to be asked to do just one thing. But Peter would enforce this pretty strictly. He'd basically say: 'I will not talk to you about anything else except for this one thing that I've assigned to you. I don't want to hear about how great you're doing in this other area. Just focus until you conquer this one problem.'... The insight behind this is that most people will solve problems that they understand how to solve. Roughly speaking, they will solve B+ problems instead of A+ problems. A+ problems are high-impact problems for your company but they're difficult--you don't wake up in the morning with a solution to them, so you tend to procrastinate... If you have a company that's always solving B+ problems, you'll never create the breakthrough idea because no one is spending 100% of their time banging their head against the wall every day until they solve it" Video source: Y Combinator (2014)

Startup Archive

108,875 views • 1 year ago

"What I always wonder with the club when they put these things out - and I've said this to them - is what problem are they trying to solve? I think they're trying to encourage: if you can't go, you sell it back to the club - but people don't trust that system." 🎥 Talking Reds

"What I always wonder with the club when they put these things out - and I've said this to them - is what problem are they trying to solve? I think they're trying to encourage: if you can't go, you sell it back to the club - but people don't trust that system." 🎥 Talking Reds

The Anfield Wrap

117,813 views • 1 year ago

How to match the complexity of the problem you want to solve with the proper model. You want an inference router. In the video, I show you how simple and powerful this is. After this, you'll never talk directly to a model ever again.

How to match the complexity of the problem you want to solve with the proper model. You want an inference router. In the video, I show you how simple and powerful this is. After this, you'll never talk directly to a model ever again.

Santiago

11,903 views • 1 month ago

Naval Ravikant on the importance of hiring high-agency people Naval defines agency as: “People who just solve problems without even being asked to solve the problem—they identify the problem, they go solve it, they don’t even necessarily have to update you every step of the way, they’re not asking silly questions, and they’re just coming up with solutions.” He believes this is important because “building a startup is an infinite set of problems that are being thrown at you.” And there comes a day where you can’t even look at every problem your company is facing—let alone solve every one of them. He cites the Vinod Khosla aphorism: "The team you build is the company you build, not the plan you make.” And your ability to solve problems is based entirely on how many problem-solvers you have at your company. As Naval puts it: “If you have somebody who takes 10% of your time and management to solve problems, you can only have 10 of those people working with you. But if somebody takes 5%, you can have 20 of those people.” When building Airchat and AngelList, he thought of each team as a Navy Seal team: “Everyone is just really good at what they do. They know their job. They do it. They don’t complain. They’re not egotistical about it. And if they have to constantly be corrected, led around by the nose, you have to clean up after them, or you question their judgement, it’s not going to work out.”

Naval Ravikant on the importance of hiring high-agency people Naval defines agency as: “People who just solve problems without even being asked to solve the problem—they identify the problem, they go solve it, they don’t even necessarily have to update you every step of the way, they’re not asking silly questions, and they’re just coming up with solutions.” He believes this is important because “building a startup is an infinite set of problems that are being thrown at you.” And there comes a day where you can’t even look at every problem your company is facing—let alone solve every one of them. He cites the Vinod Khosla aphorism: "The team you build is the company you build, not the plan you make.” And your ability to solve problems is based entirely on how many problem-solvers you have at your company. As Naval puts it: “If you have somebody who takes 10% of your time and management to solve problems, you can only have 10 of those people working with you. But if somebody takes 5%, you can have 20 of those people.” When building Airchat and AngelList, he thought of each team as a Navy Seal team: “Everyone is just really good at what they do. They know their job. They do it. They don’t complain. They’re not egotistical about it. And if they have to constantly be corrected, led around by the nose, you have to clean up after them, or you question their judgement, it’s not going to work out.”

Startup Archive

552,771 views • 2 years ago

Right now, I have a few dozen tools to automate my life. Every single thing that I do more than once every week is now a tool: either a Claude Code skill, a scheduled workflow, or an application. • I spend the time doing this once. • I schedule it. • I forget about it (or at least, try to) CREAO is one of the platforms I've used extensively for building some of these automations. They are partnering with me on this post. They are one of the only platforms where you can go from a conversation to a scheduled agent that quickly. This is how it works: 1. You describe the problem you want to solve 2. CREAO's agent builds the logic and executes it 3. You can iterate with the agent to improve the solution 4. When you are done, you turn that solution into a mini app 5. You can schedule that app to run any time you need it Something important: These agents are deterministic. They don't use an LLM to generate answers, so they will always return the same output given the same input. This is critical: when you're automating something that runs every Monday at 9 am, you need it to return the same result every time, not a "creative" answer.

Right now, I have a few dozen tools to automate my life. Every single thing that I do more than once every week is now a tool: either a Claude Code skill, a scheduled workflow, or an application. • I spend the time doing this once. • I schedule it. • I forget about it (or at least, try to) CREAO is one of the platforms I've used extensively for building some of these automations. They are partnering with me on this post. They are one of the only platforms where you can go from a conversation to a scheduled agent that quickly. This is how it works: 1. You describe the problem you want to solve 2. CREAO's agent builds the logic and executes it 3. You can iterate with the agent to improve the solution 4. When you are done, you turn that solution into a mini app 5. You can schedule that app to run any time you need it Something important: These agents are deterministic. They don't use an LLM to generate answers, so they will always return the same output given the same input. This is critical: when you're automating something that runs every Monday at 9 am, you need it to return the same result every time, not a "creative" answer.

Santiago

13,712 views • 3 months ago

3/ Companies peddle the BS that advertising data is 'anonymous' They want to keep you in the dark. Because, behind the scenes...your weather app is a blinking tracking beacon. And then players like PenLink turn that beacon into a data flow. And sell it to governments.

3/ Companies peddle the BS that advertising data is 'anonymous' They want to keep you in the dark. Because, behind the scenes...your weather app is a blinking tracking beacon. And then players like PenLink turn that beacon into a data flow. And sell it to governments.

John Scott-Railton

37,960 views • 3 months ago

from their convo, you can literally see how close they are. haruto is always talking about jeongwoo, and they understand each other so well. i really need them to have a show like this together where they can just sit and talk about random things #하루토 #박정우

from their convo, you can literally see how close they are. haruto is always talking about jeongwoo, and they understand each other so well. i really need them to have a show like this together where they can just sit and talk about random things #하루토 #박정우

jeongwooxharuto | 박정우 x 하루토

12,411 views • 1 month ago

"Do you want to know how difficult it is to confront FPV drones on the battlefield? And how a child controlling one remotely can take out a soldier with long military experience? Watch this video… They decided to try it themselves and were shocked—they didn’t know how to deal with them. That’s why the war in Ukraine is so tragic."

"Do you want to know how difficult it is to confront FPV drones on the battlefield? And how a child controlling one remotely can take out a soldier with long military experience? Watch this video… They decided to try it themselves and were shocked—they didn’t know how to deal with them. That’s why the war in Ukraine is so tragic."

War Archive Clips

578,848 views • 3 months ago

This is probably the biggest issue that people have with their lives A negative mindset, fed by negative talking and thinking. You will find it very hard to positively improve your life if you continue to do this. They perpetuate negative circumstances by talking about them as if they are current. You may have experienced something in the past, but referring to it and giving it attention and energy is only strengthening it. Do away with the idea that you must be "logical" and "normal" with how you talk and live. Instead, put your mind and energy into the reality you wish to experience, and your life will begin to align around that.

This is probably the biggest issue that people have with their lives A negative mindset, fed by negative talking and thinking. You will find it very hard to positively improve your life if you continue to do this. They perpetuate negative circumstances by talking about them as if they are current. You may have experienced something in the past, but referring to it and giving it attention and energy is only strengthening it. Do away with the idea that you must be "logical" and "normal" with how you talk and live. Instead, put your mind and energy into the reality you wish to experience, and your life will begin to align around that.

⚡️🌞 Sol Brah 🌞🐬

15,529 views • 8 months ago

Forget about complicated Excel formulas. Leave it in 2008. Using this AI platform, you can create presentations with interactive charts just by using a prompt. Here is how you can turn any data into beautiful visual presentations: Gamma — the all-in-one platform to create social media posts, presentations, landing pages, and documents in less than a minute using AI. Forget about manually designing; this tool is a game changer. Let me show you how it works in action: I went to [ and I asked it: "Create a presentation with charts showing New York immigration data and its impact on music culture" In less than a minute, Gamma created an 8-page beautifully looking presentation. It works especially for people who need to present ideas fast and do not have time. Literally prompt and watch it cook. Not only that, it also broke down the immigration population by division percentage. So it's not only making the presentations look good but also making them practical. Over 50 million users are using Gamma every day to be 1% better at their work. And if you're still not on board, now is your time. Give Gamma a try for free today:

Forget about complicated Excel formulas. Leave it in 2008. Using this AI platform, you can create presentations with interactive charts just by using a prompt. Here is how you can turn any data into beautiful visual presentations: Gamma — the all-in-one platform to create social media posts, presentations, landing pages, and documents in less than a minute using AI. Forget about manually designing; this tool is a game changer. Let me show you how it works in action: I went to [ and I asked it: "Create a presentation with charts showing New York immigration data and its impact on music culture" In less than a minute, Gamma created an 8-page beautifully looking presentation. It works especially for people who need to present ideas fast and do not have time. Literally prompt and watch it cook. Not only that, it also broke down the immigration population by division percentage. So it's not only making the presentations look good but also making them practical. Over 50 million users are using Gamma every day to be 1% better at their work. And if you're still not on board, now is your time. Give Gamma a try for free today:

Alif Hossain

34,541 views • 10 months ago

Web scraping is a critical skill, and yet nobody talks about it. How do you think companies are training their Large Language Models? Where do you think the data comes from? But web scraping goes beyond all of that. Imagine giving an AI agent access to any public online data in real time! I like to call this "web-scrapping on demand", and I'm pretty sure it's going to unlock unlimited power for AI applications. I recorded a quick video to show you how you can do this using Apify. I've talked about them before, and they are collaborating with me on this post. They have one of the best open-source web scraping and browser automation libraries out there: But it gets much better than this! You can use MCP to connect your AI Agents and applications to the Apify platform and use any specialized actor on demand to scrape and process online data. In the video, I used Cursor to scrape LinkedIn posts with the words "Machine Learning" in real time. Worked like a charm with no code needed! Here is a link to the platform: Think about this: You can now feed your AI applications with any public data on demand! We aren't ready for what's coming.

Web scraping is a critical skill, and yet nobody talks about it. How do you think companies are training their Large Language Models? Where do you think the data comes from? But web scraping goes beyond all of that. Imagine giving an AI agent access to any public online data in real time! I like to call this "web-scrapping on demand", and I'm pretty sure it's going to unlock unlimited power for AI applications. I recorded a quick video to show you how you can do this using Apify. I've talked about them before, and they are collaborating with me on this post. They have one of the best open-source web scraping and browser automation libraries out there: But it gets much better than this! You can use MCP to connect your AI Agents and applications to the Apify platform and use any specialized actor on demand to scrape and process online data. In the video, I used Cursor to scrape LinkedIn posts with the words "Machine Learning" in real time. Worked like a charm with no code needed! Here is a link to the platform: Think about this: You can now feed your AI applications with any public data on demand! We aren't ready for what's coming.

Santiago

101,473 views • 11 months ago

this is wild this AR app that lets you talk to a book and turn it into a quiz..

this is wild this AR app that lets you talk to a book and turn it into a quiz..

Aurelien

32,999 views • 8 months ago