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.

So dang enchanted with the Gemini 3.0 Flash support in Browserbase, especially for tedious data tasks! Check out the video below - Gemini navigates to the USGS website for seismic data, filters on 3D Multichannel Seismic datasets, adds all of the options for California, and then downloads as a... show more

👩‍💻 Paige Bailey

76,105 subscribers

25,718 views • 10 months ago •via X (Twitter)

Science & Technology Education

Anya Rossi• Live Now

Private livecam show

9 Comments

👩‍💻 Paige Bailey10 months ago

Since @Antigravity has computer use and screen understanding built in, it can also handle these kinds of tasks - and much more flexibly. Here it is going to every one of the extracted URLs, figuring out which link should be clicked to download the seismic, and then getting the extracted data prepared for visualization. I love inspecting the reasoning traces, too: Goal: Visit the following three URLs and report back on the availability and size of "dataset" files: 1. 2. 3. Compare these to the first one (p-04-11-cc_eb). Are there always small navigation files? Is the seismic data always large? Is there a common format for the "dataset" (e.g. CSV, Shapefile)?

👩‍💻 Paige Bailey10 months ago

Prompt: Download as a CSV List all of the 3D Multichannel Seismic datasets in California from 🔗

Khanhvksnb.eth | Ⓜ️Ⓜ️T10 months ago

@browserbase @USGS dang, feels like gemini's on a whole new level of boring data tasks. tbh, does the flash support work seamlessly for real-time updates or just static downloads? kinda keen to see how it shakes up data science workflows. ngmi if i don't keep up with this wizardry!

Joshua Poddoku10 months ago

@browserbase @USGS This looks slick for data tasks

AI~rtesan10 months ago

@browserbase @USGS Oooh, the @GoogleColab extra flavor is a nice topping!🍨

Jack Crabtree10 months ago

@browserbase @USGS Browser agents scraping public data look flashy until sites change layouts and everything breaks. Stick to APIs for reliable RevOps feeds.

Karan Jagtiani10 months ago

@browserbase @USGS Looks user-friendly for data tasks. Curious how it handles larger datasets

Himanshu Kumar10 months ago

@browserbase @USGS I've observed Gemini's Flash support assisting with complex data extraction tasks.

na10 months ago

@browserbase @USGS Is browserbase better than Gemini CLI controlling Chrome for this use?

Related Videos

Automatic code generation and data visualization for #EarthEngine and #geemap using #Gemini 🚀 Explore the power of cloud computing and #geospatial big data with minimal coding 🌎. Gemini can interpret the data on the map and iteratively improve the visualization. Check out the notebook:

Automatic code generation and data visualization for #EarthEngine and #geemap using #Gemini 🚀 Explore the power of cloud computing and #geospatial big data with minimal coding 🌎. Gemini can interpret the data on the map and iteratively improve the visualization. Check out the notebook:

Qiusheng Wu

22,387 views • 2 years ago

I made an AI data science team with Python. I'm excited to share my new project, the AI Data Science Team, a Python package for creating AI co-pilots and agents to assist in common data science tasks. I walk you through the GitHub page, showcasing the progress and potential of this project. I'll explain how these agents can streamline tasks like code generation and modeling pipeline creation, making data science projects more efficient. Introduction to AI Data Science Team Project Integrating Human-in-the-Loop Example: Accessing Agents for Feature Engineering Setting up Feature Engineering Agent Code On GitHub: Python AI-Tips Newsletter (stay updated as I build this project in the open):

I made an AI data science team with Python. I'm excited to share my new project, the AI Data Science Team, a Python package for creating AI co-pilots and agents to assist in common data science tasks. I walk you through the GitHub page, showcasing the progress and potential of this project. I'll explain how these agents can streamline tasks like code generation and modeling pipeline creation, making data science projects more efficient. Introduction to AI Data Science Team Project Integrating Human-in-the-Loop Example: Accessing Agents for Feature Engineering Setting up Feature Engineering Agent Code On GitHub: Python AI-Tips Newsletter (stay updated as I build this project in the open):

Matt Dancho (Business Science)

11,799 views • 1 year ago

📒 Automating market research with Google AI Studio Gemini Search grounding + JSON mode this evening: 1⃣ Use Gemini 1.5 Flash + Search grounding to pull recent data about a list of companies, their product offerings, valuations, more. 2⃣ Use JSON mode with Gemini 1.5 Flash-8B to extract out specific details, for ~$0. 3⃣ Write data back to Google Sheets from Colaboratory:

📒 Automating market research with Google AI Studio Gemini Search grounding + JSON mode this evening: 1⃣ Use Gemini 1.5 Flash + Search grounding to pull recent data about a list of companies, their product offerings, valuations, more. 2⃣ Use JSON mode with Gemini 1.5 Flash-8B to extract out specific details, for ~$0. 3⃣ Write data back to Google Sheets from Colaboratory:

👩‍💻 Paige Bailey

43,383 views • 1 year ago

Economics and Data Editor, Ed Conway, says the global economic impact of the situation in the Middle East could be ‘seismic’ ⬇️ Latest:

Economics and Data Editor, Ed Conway, says the global economic impact of the situation in the Middle East could be ‘seismic’ ⬇️ Latest:

Sky News

19,568 views • 8 months ago

Gemini 3.5 Flash can process complex visual data and translate it into functional, interactive code. Watch Gemini analyze lighting from a reference image, and build an interactive 3D visualizer to preview the setup.

Gemini 3.5 Flash can process complex visual data and translate it into functional, interactive code. Watch Gemini analyze lighting from a reference image, and build an interactive 3D visualizer to preview the setup.

Google Gemini

1,129,301 views • 5 months ago

Gemini 3.5 flash + Gemini managed agents api just audited a real megatron-lm ci failure inside Eigent. root cause in minutes! watch the handoff: coordinator agent plans the audit, developer agent loads the ml-failure-audit skill and gathers the evidence, then gemini agent steps in as a remote sub-agent for the heavy reasoning. gemini managed agents api and gemini 3.5 flash now live on our open-source cowork Eigent! Google AI Developers Google AI Studio

Gemini 3.5 flash + Gemini managed agents api just audited a real megatron-lm ci failure inside Eigent. root cause in minutes! watch the handoff: coordinator agent plans the audit, developer agent loads the ml-failure-audit skill and gathers the evidence, then gemini agent steps in as a remote sub-agent for the heavy reasoning. gemini managed agents api and gemini 3.5 flash now live on our open-source cowork Eigent! Google AI Developers Google AI Studio

Eigent AI

17,260 views • 5 months ago

Today we're launching Amplitude MCP MCP was made for data analytics. Ask a vague question and have an AI model iteratively query Amplitude to find you insights. Amplitude MCP exposes all of Amplitude's functionality so an agent can interact with it directly. It allows you to use Amplitude without having to know anything about Amplitude's user interface or your data taxonomy. It uses the intelligence of off the shelf AI agents, including Claude, Gemini, and Cursor, to query Amplitude data you already have to find answers. It leverages all of your existing Amplitude data and how it's used to navigate through large complex behavioral datasets. You can have conversations about your product data directly inside any AI model with MCP. It can create Amplitude dashboards and notebooks with what it finds and even make recommendations. This is the best MCP use case I've ever seen. Watching an AI agent think, reason, query Amplitude, and then repeat that process iteratively is magical. See it in action:

Today we're launching Amplitude MCP MCP was made for data analytics. Ask a vague question and have an AI model iteratively query Amplitude to find you insights. Amplitude MCP exposes all of Amplitude's functionality so an agent can interact with it directly. It allows you to use Amplitude without having to know anything about Amplitude's user interface or your data taxonomy. It uses the intelligence of off the shelf AI agents, including Claude, Gemini, and Cursor, to query Amplitude data you already have to find answers. It leverages all of your existing Amplitude data and how it's used to navigate through large complex behavioral datasets. You can have conversations about your product data directly inside any AI model with MCP. It can create Amplitude dashboards and notebooks with what it finds and even make recommendations. This is the best MCP use case I've ever seen. Watching an AI agent think, reason, query Amplitude, and then repeat that process iteratively is magical. See it in action:

Spenser Skates

32,477 views • 1 year ago

Google's best model is actually hidden Gemini 2.5 Pro with Web Grounding is incredible and only available in AI Studio You get free access to the most powerful model with real-time data: - Access AI Studio - Select Gemini 2.5 Pro Preview - Activate Grounding with Search Gemini will search the data, think, then give you the answers with the sources.

Google's best model is actually hidden Gemini 2.5 Pro with Web Grounding is incredible and only available in AI Studio You get free access to the most powerful model with real-time data: - Access AI Studio - Select Gemini 2.5 Pro Preview - Activate Grounding with Search Gemini will search the data, think, then give you the answers with the sources.

Paul Couvert

150,053 views • 1 year ago

Learning from Human Demonstrations: Show the Robot How to Act! The pipeline is very similar to older experiments using Gemini & pi0 with LeRobot. Pi-zero runs locally, while Gemini Flash generates the affordances and the high-level task. (More details are in the thread.) The new component is learning from demonstrations via Gemini 2.5 Pro. I capture a video while demoing & take one of the last frames. Gemini 2.5 Pro then extracts the instructions & passes them to Gemini Flash to process the scene. The fun part is that there's no fancy insight that came from me; other than the days spent figuring out the right prompts. It's the bitter lesson hitting you in the face -> Enhanced Gemini capabilities make this possible. For example, Gemini Flash cannot do Russian doll stacking, but Gemini 2.5 Pro can do it consistently. The current limitation is low-level manipulation: - As you can see, I'm aligning the objects so they are easy to grasp using the same technique from the training data. I couldn't get Gemini Flash to consistently output an accurate grasping angle, and Gemini 1.5 Pro was too expensive and slow for real-time deployment. - Getting a symmetrical gripper should also help a lot. Adding rubber to the tips would probably also help prevent objects from slipping. Collecting & curating the data was the most time consuming & labor intensive part. Next, to improve low-level manipulation and make the system more real-time, I'm shifting to focus more on sims & synthetic data. This aligns better with my core competence. I'm open to tips and suggestions.

Learning from Human Demonstrations: Show the Robot How to Act! The pipeline is very similar to older experiments using Gemini & pi0 with LeRobot. Pi-zero runs locally, while Gemini Flash generates the affordances and the high-level task. (More details are in the thread.) The new component is learning from demonstrations via Gemini 2.5 Pro. I capture a video while demoing & take one of the last frames. Gemini 2.5 Pro then extracts the instructions & passes them to Gemini Flash to process the scene. The fun part is that there's no fancy insight that came from me; other than the days spent figuring out the right prompts. It's the bitter lesson hitting you in the face -> Enhanced Gemini capabilities make this possible. For example, Gemini Flash cannot do Russian doll stacking, but Gemini 2.5 Pro can do it consistently. The current limitation is low-level manipulation: - As you can see, I'm aligning the objects so they are easy to grasp using the same technique from the training data. I couldn't get Gemini Flash to consistently output an accurate grasping angle, and Gemini 1.5 Pro was too expensive and slow for real-time deployment. - Getting a symmetrical gripper should also help a lot. Adding rubber to the tips would probably also help prevent objects from slipping. Collecting & curating the data was the most time consuming & labor intensive part. Next, to improve low-level manipulation and make the system more real-time, I'm shifting to focus more on sims & synthetic data. This aligns better with my core competence. I'm open to tips and suggestions.

Shreyas Gite

22,555 views • 1 year ago

We (Pantheon) have been working on a robotics data quality pipeline that uncovered a series of major problems in public robotics datasets, especially for world modeling. To improve the quality of data available to open-source robotics, we're publishing annotations for four of the most popular datasets. Some examples of issues, and our report 🧵

We (Pantheon) have been working on a robotics data quality pipeline that uncovered a series of major problems in public robotics datasets, especially for world modeling. To improve the quality of data available to open-source robotics, we're publishing annotations for four of the most popular datasets. Some examples of issues, and our report 🧵

calix

183,598 views • 1 month ago

#WATCH | Delhi | UIDAI Chairperson and World Bank Executive Director Neelkanth Mishra says, "...What is most critical is that 99% of the data in corporations is untouched by AI. None of the companies wants to share this data with AI models because once you've shared it, it's gone. And if you don't have that, then how do you get solutions?.. There is a complete market failure in the data market... India's Digital Personal Data Protection (DPDP) law, the data architecture which is based on DEPA (Data Empowerment and Protection Architecture), is a remarkable model for enabling data sharing where the owner of the data still retains control even as it makes it available for certain usage..."

#WATCH | Delhi | UIDAI Chairperson and World Bank Executive Director Neelkanth Mishra says, "...What is most critical is that 99% of the data in corporations is untouched by AI. None of the companies wants to share this data with AI models because once you've shared it, it's gone. And if you don't have that, then how do you get solutions?.. There is a complete market failure in the data market... India's Digital Personal Data Protection (DPDP) law, the data architecture which is based on DEPA (Data Empowerment and Protection Architecture), is a remarkable model for enabling data sharing where the owner of the data still retains control even as it makes it available for certain usage..."

ANI

36,038 views • 1 month ago

a Stanford professor used JEV to build a data science system that analyzes 40.000.000.000 data points every 15 min his system can process massive datasets and automatically decide which results deserve deeper analysis. the professor spent more than a year developing the system with his research team. the first version relied heavily on LLMs to interpret every result. it worked, but analyzing millions of records created unnecessary latency and huge inference costs. then they redesigned the decision layer around Jev. instead of generating explanations for every data point, Jev evaluates the results and decides what the system should investigate next. I found a breakdown of the pipeline they used and the way Jev filters the data is surprisingly clever. I’ll show you how it works next so stay tuned if you want to see what Jev Engineering looks like inside a real data science system. would you let an AI decide which patterns in your data are worth investigating?

a Stanford professor used JEV to build a data science system that analyzes 40.000.000.000 data points every 15 min his system can process massive datasets and automatically decide which results deserve deeper analysis. the professor spent more than a year developing the system with his research team. the first version relied heavily on LLMs to interpret every result. it worked, but analyzing millions of records created unnecessary latency and huge inference costs. then they redesigned the decision layer around Jev. instead of generating explanations for every data point, Jev evaluates the results and decides what the system should investigate next. I found a breakdown of the pipeline they used and the way Jev filters the data is surprisingly clever. I’ll show you how it works next so stay tuned if you want to see what Jev Engineering looks like inside a real data science system. would you let an AI decide which patterns in your data are worth investigating?

deezzex

175,647 views • 1 month ago

We successfully launched a sensor into a tornado with good data back! Here is the first drop of the 3D data!!!! Check out more of our science data on (we are working on better analytics for the site.) This dataset is from the Blue Rapids KS tornado on 5/18/26

We successfully launched a sensor into a tornado with good data back! Here is the first drop of the 3D data!!!! Check out more of our science data on (we are working on better analytics for the site.) This dataset is from the Blue Rapids KS tornado on 5/18/26

Will Clay

16,259 views • 5 months ago

PhD Students – How to automatically extract data from papers for your literature review? Extracting relevant data from papers is challenging. However, this process can be automated. Meet AnswerThis – a tool that extracts data in seconds. Here is how it works. 1. Go to and log in. 2. After logging in, click on 𝐸𝑥𝑡𝑟𝑎𝑐𝑡 𝑑𝑎𝑡𝑎. 3. Then click on 𝑈𝑝𝑙𝑜𝑎𝑑 𝑃𝐷𝐹 and upload your papers. 4. These are the papers from which you want to extract data. 5. After uploading papers, select data you want to extract. 6. The predefined options are - Key findings - Research gaps - Methodology - Limitations - Future work - Contributions - Practical implications 7. You can also extract custom data e.g., dataset used. 8. For example, I want to extract methodology used in these papers. 9. I selected 𝑀𝑒𝑡ℎ𝑜𝑑𝑜𝑙𝑜𝑔𝑦 and clicked on 𝐴𝑑𝑑 𝐶𝑜𝑙𝑢𝑚𝑛. 10. AnswerThis extract data about methodology used in the papers. 11. You can change data view from normal to Table View. 12. For this, scroll back to top and click on 𝑇𝑎𝑏𝑙𝑒 𝑉𝑖𝑒𝑤. 13. Now for instance, you want to extract more data from these papers. 14. Go back to the top and click on 𝐸𝑥𝑡𝑟𝑎𝑐𝑡 𝑑𝑎𝑡𝑎. 15. Select the data type you want to extract. 16. For example, I want to extract data about future work. 17. So I click on 𝐹𝑢𝑡𝑢𝑟𝑒 𝑊𝑜𝑟𝑘 and then clicked on 𝐴𝑑𝑑 𝑐𝑜𝑙𝑢𝑚𝑛. 18. AnswerThis extracted data about future work from the papers. 19. After extracting the desired data, you can export it. 20. Select the data you want to extract. 21. Then click on 𝐸𝑥𝑝𝑜𝑟𝑡 𝑑𝑎𝑡𝑎. 22. Your data will be exported in CSV format. You can then analyze this data for your literature review. Try AnswerThis today: Anything you'd like to add?

PhD Students – How to automatically extract data from papers for your literature review? Extracting relevant data from papers is challenging. However, this process can be automated. Meet AnswerThis – a tool that extracts data in seconds. Here is how it works. 1. Go to and log in. 2. After logging in, click on 𝐸𝑥𝑡𝑟𝑎𝑐𝑡 𝑑𝑎𝑡𝑎. 3. Then click on 𝑈𝑝𝑙𝑜𝑎𝑑 𝑃𝐷𝐹 and upload your papers. 4. These are the papers from which you want to extract data. 5. After uploading papers, select data you want to extract. 6. The predefined options are - Key findings - Research gaps - Methodology - Limitations - Future work - Contributions - Practical implications 7. You can also extract custom data e.g., dataset used. 8. For example, I want to extract methodology used in these papers. 9. I selected 𝑀𝑒𝑡ℎ𝑜𝑑𝑜𝑙𝑜𝑔𝑦 and clicked on 𝐴𝑑𝑑 𝐶𝑜𝑙𝑢𝑚𝑛. 10. AnswerThis extract data about methodology used in the papers. 11. You can change data view from normal to Table View. 12. For this, scroll back to top and click on 𝑇𝑎𝑏𝑙𝑒 𝑉𝑖𝑒𝑤. 13. Now for instance, you want to extract more data from these papers. 14. Go back to the top and click on 𝐸𝑥𝑡𝑟𝑎𝑐𝑡 𝑑𝑎𝑡𝑎. 15. Select the data type you want to extract. 16. For example, I want to extract data about future work. 17. So I click on 𝐹𝑢𝑡𝑢𝑟𝑒 𝑊𝑜𝑟𝑘 and then clicked on 𝐴𝑑𝑑 𝑐𝑜𝑙𝑢𝑚𝑛. 18. AnswerThis extracted data about future work from the papers. 19. After extracting the desired data, you can export it. 20. Select the data you want to extract. 21. Then click on 𝐸𝑥𝑝𝑜𝑟𝑡 𝑑𝑎𝑡𝑎. 22. Your data will be exported in CSV format. You can then analyze this data for your literature review. Try AnswerThis today: Anything you'd like to add?

Faheem Ullah

21,390 views • 1 year ago

Excited to share that maplibre-gl-lidar v0.6.0 is now available! This release adds support for the Entwine Point Tile (EPT) format ( making it possible to stream and visualize billions of LiDAR points directly in the browser with ease and performance. The demo below showcases the visualization of an incredible 184 billion LiDAR points! Live Demo Viewer: GitHub Repository: Full Video Tutorial: Huge thanks to U.S. Geological Survey (USGS) and the 3DEP Program ( for acquiring nationwide LiDAR data, Hobu, Inc. for processing the point clouds into Entwine format ( and the Amazon Web Services (AWS) Open Data Program ( for hosting this massive and continually growing collection of over 350 TB of EPT data with over 75 trillion points. #lidar #opensource #opendata #geospatial #maplibre #deckgl

Excited to share that maplibre-gl-lidar v0.6.0 is now available! This release adds support for the Entwine Point Tile (EPT) format ( making it possible to stream and visualize billions of LiDAR points directly in the browser with ease and performance. The demo below showcases the visualization of an incredible 184 billion LiDAR points! Live Demo Viewer: GitHub Repository: Full Video Tutorial: Huge thanks to U.S. Geological Survey (USGS) and the 3DEP Program ( for acquiring nationwide LiDAR data, Hobu, Inc. for processing the point clouds into Entwine format ( and the Amazon Web Services (AWS) Open Data Program ( for hosting this massive and continually growing collection of over 350 TB of EPT data with over 75 trillion points. #lidar #opensource #opendata #geospatial #maplibre #deckgl

Qiusheng Wu

13,095 views • 10 months ago

For me, our Analyst agent in Microsoft 365 Copilot is like having a skilled data analyst on hand — all the time. It generates insights and helps you visualize complex data. Here's a quick demo on how it can make sense of marketing campaign data and help you get more done at work.

For me, our Analyst agent in Microsoft 365 Copilot is like having a skilled data analyst on hand — all the time. It generates insights and helps you visualize complex data. Here's a quick demo on how it can make sense of marketing campaign data and help you get more done at work.

Satya Nadella

351,274 views • 1 year ago

💡Why Vana? “Ultimately, AI models are only as good as their training data. So if you want to build the best AI, you need the best data.” shares the vision behind Vana’s mission, centered on DataDAOs that aggregate specific datasets, reward data owners for their contributions and level up AI models with high-quality data🏆

💡Why Vana? “Ultimately, AI models are only as good as their training data. So if you want to build the best AI, you need the best data.” shares the vision behind Vana’s mission, centered on DataDAOs that aggregate specific datasets, reward data owners for their contributions and level up AI models with high-quality data🏆

vana

108,377 views • 1 year ago