Video wird geladen...

Video konnte nicht geladen werden

Beim Laden dieses Videos ist ein Problem aufgetreten. Dies könnte an einem vorübergehenden Netzwerkproblem liegen oder das Video ist möglicherweise nicht verfügbar.

We're making Extract even better! It's now twice as fast, more accurate, and a fraction of the cost. What's new: → Faster v4 model with better schema adherence & up to 99% citation coverage → Flat 2¢/page pricing, all-in. No added parse costs. → Upcoming MCP tools to help... show more

Adit

7,295 subscribers

19,202 Aufrufe • vor 4 Tagen •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

21 Kommentare

Profilbild von Arjun Chaliha

Arjun Chalihavor 4 Tagen

@reductoai @TheOtherGandhi_ Lfg you guys are helping us cook

Profilbild von Adit

Aditvor 4 Tagen

@reductoai @TheOtherGandhi_ 🫡 we'll keep pushing

Profilbild von Adit

Aditvor 4 Tagen

You can try the v4 model in studio today: If you're using another document processing solution, we're offering up to $5000 in migration credits:

Profilbild von adel 🌟

adel 🌟vor 4 Tagen

@donaldwu_ @reductoai @TheOtherGandhi_ my steak too juicy my lobster too buttery

Profilbild von Donald

Donaldvor 4 Tagen

Twice as fast has to be the biggest part for us here. There are so many real-time use cases that I think will be a huge unlock from this release. If you haven't given /extract a try, now is a great time.

Profilbild von Adit

Aditvor 4 Tagen

@reductoai @TheOtherGandhi_ I nearly leaked the next extract launch as a reply to this comment But instead I will stick to this vaguepost

Profilbild von Apurv Gandhi

Apurv Gandhivor 4 Tagen

@reductoai I think it’s gotten even faster from the time this video was filmed :)

Profilbild von palak

palakvor 4 Tagen

@reductoai @TheOtherGandhi_ the best just keeps getting better 🚀🚀

Profilbild von Vineeth

Vineethvor 4 Tagen

@reductoai @TheOtherGandhi_ Interesting 🤔

Profilbild von Jeremy Bosma

Jeremy Bosmavor 4 Tagen

@reductoai @TheOtherGandhi_ MCP tools turn extraction from a feature into infrastructure.

Profilbild von meg hildreth

meg hildrethvor 4 Tagen

@reductoai @TheOtherGandhi_ excited about your mcp - we should talk getting you guys into the marketplaces + mcp analytics bc @manufact can help with that ;)

Profilbild von mangi

mangivor 4 Tagen

@reductoai @TheOtherGandhi_ good stuff, lets connect!

Profilbild von Hatef | AGINAZ

Hatef | AGINAZvor 4 Tagen

@reductoai @TheOtherGandhi_ Schema adherence is the bigger win here. I keep seeing extraction look valid even when the source does not support a required field.

Profilbild von Sebastian Bugal

Sebastian Bugalvor 4 Tagen

@reductoai @TheOtherGandhi_ Does this apply to extract and deep extract?

Profilbild von Adit

Aditvor 4 Tagen

@reductoai @TheOtherGandhi_ deep extract is also shifting to simpler/lower pricing! it's 4c/page vs 2c for base extract

Profilbild von Sebastian Bugal

Sebastian Bugalvor 3 Tagen

@reductoai @TheOtherGandhi_ Ok sick! btw love what yall did with deep extract fixed 90% of the problems I had with extract

Profilbild von Adit

Aditvor 3 Tagen

@reductoai @TheOtherGandhi_ cc: @ArthBohra

Profilbild von Tony Hu

Tony Huvor 3 Tagen

The upcoming schema tools sound especially useful. A faster model helps, but making the setup easier can matter just as much. Clear citation coverage also makes the output easier to check.

Profilbild von Sylvain Perrier

Sylvain Perriervor 3 Tagen

@reductoai @TheOtherGandhi_ The operating leverage is the real edge here: predictable per-page pricing plus stronger schema adherence turns extraction from a bespoke task into a repeatable workflow.

Profilbild von Jason傑森 🇭🇰 | 🛠️

Jason傑森 🇭🇰 | 🛠️vor 3 Tagen

@reductoai @TheOtherGandhi_ 这么多好的地方价格两美分划算吗

Profilbild von Chen

Chenvor 4 Tagen

@reductoai @TheOtherGandhi_ SaaS problems, classic

Ähnliche Videos

After the release of Parse v2, Extract is also getting an upgrade — 𝗶𝗻𝘁𝗿𝗼𝗱𝘂𝗰𝗶𝗻𝗴 𝗘𝘅𝘁𝗿𝗮𝗰𝘁 𝘃𝟮! 🎉 We've been reworking the experience from the ground up to make document extraction more powerful and easier to use than ever. Here's what's new: ✦ 𝗦𝗶𝗺𝗽𝗹𝗶𝗳𝗶𝗲𝗱 𝘁𝗶𝗲𝗿𝘀: we've replaced modes with cleaner, more intuitive tiers. (And stay tuned: agentic plus is coming to Extract too, very soon.) ✦ 𝗣𝗿𝗲-𝘀𝗮𝘃𝗲𝗱 𝗲𝘅𝘁𝗿𝗮𝗰𝘁 𝗰𝗼𝗻𝗳𝗶𝗴𝘂𝗿𝗮𝘁𝗶𝗼𝗻𝘀: load your saved extraction configs directly, so you can skip the setup and get straight to extracting. ✦ 𝗖𝗼𝗻𝗳𝗶𝗴𝘂𝗿𝗮𝗯𝗹𝗲 𝗱𝗼𝗰𝘂𝗺𝗲𝗻𝘁 𝗽𝗮𝗿𝘀𝗶𝗻𝗴: now you can control how your documents get parsed before extraction, giving you more flexibility and better results end to end. And for those who need a transition period: Extract v1 will remain accessible via the UI under 'Settings → General' for a limited time. Try Extract v2 today →

After the release of Parse v2, Extract is also getting an upgrade — 𝗶𝗻𝘁𝗿𝗼𝗱𝘂𝗰𝗶𝗻𝗴 𝗘𝘅𝘁𝗿𝗮𝗰𝘁 𝘃𝟮! 🎉 We've been reworking the experience from the ground up to make document extraction more powerful and easier to use than ever. Here's what's new: ✦ 𝗦𝗶𝗺𝗽𝗹𝗶𝗳𝗶𝗲𝗱 𝘁𝗶𝗲𝗿𝘀: we've replaced modes with cleaner, more intuitive tiers. (And stay tuned: agentic plus is coming to Extract too, very soon.) ✦ 𝗣𝗿𝗲-𝘀𝗮𝘃𝗲𝗱 𝗲𝘅𝘁𝗿𝗮𝗰𝘁 𝗰𝗼𝗻𝗳𝗶𝗴𝘂𝗿𝗮𝘁𝗶𝗼𝗻𝘀: load your saved extraction configs directly, so you can skip the setup and get straight to extracting. ✦ 𝗖𝗼𝗻𝗳𝗶𝗴𝘂𝗿𝗮𝗯𝗹𝗲 𝗱𝗼𝗰𝘂𝗺𝗲𝗻𝘁 𝗽𝗮𝗿𝘀𝗶𝗻𝗴: now you can control how your documents get parsed before extraction, giving you more flexibility and better results end to end. And for those who need a transition period: Extract v1 will remain accessible via the UI under 'Settings → General' for a limited time. Try Extract v2 today →

LlamaIndex 🦙

15,041 Aufrufe • vor 5 Monaten

Astra is awesome, but if you're using it specifically to parse and extract a massive volume of documents, you will break the bank (10c+ per page on parsing and extraction!) We've added a native LlamaParse connector to ChatGPT that gives the model access to high-quality document OCR and extraction at the pareto frontier of accuracy and cost. It contains native access to our parse and extract endpoints. Upload your most complex documents, and you can do the following: * digitalize it into extremely accurate context for any QA task * perform large-scale extraction from a set of documents with accurate grounding and citations Offload the higher-level reasoning to Astra, leave specialized doc parsing and extraction to us. ChatGPT plugin: Come sign up here:

Astra is awesome, but if you're using it specifically to parse and extract a massive volume of documents, you will break the bank (10c+ per page on parsing and extraction!) We've added a native LlamaParse connector to ChatGPT that gives the model access to high-quality document OCR and extraction at the pareto frontier of accuracy and cost. It contains native access to our parse and extract endpoints. Upload your most complex documents, and you can do the following: * digitalize it into extremely accurate context for any QA task * perform large-scale extraction from a set of documents with accurate grounding and citations Offload the higher-level reasoning to Astra, leave specialized doc parsing and extraction to us. ChatGPT plugin: Come sign up here:

Jerry Liu

10,796 Aufrufe • vor 3 Tagen

Introducing Document Extraction as an MCP Server ✂️📑 A huge use case for AI agents is being able to extract out items from a diverse set of complex documents in a repeatable manner - whether it’s legal contracts, invoices, financial statements, passports, and more. In this case, 🚫 prompting ChatGPT/Claude is too error-prone/not systematic. 💡We’ve created a plug-and-play MCP server around our core document extraction capabilities within LlamaCloud. 1. Build an extraction agent in LlamaCloud with a standardized schema. 2. Plug it as an MCP tool within ChatGPT/Claude. 3. For your next extraction job, get the outputs in a standardized format through the tool! Huge shoutout to Tuana and Logan Markewich for working on this - this is thanks to the latest updates to our llamacloud-mcp demo repo, which supports both extraction and retrieval as MCP tools. Repo: Sign up:

Introducing Document Extraction as an MCP Server ✂️📑 A huge use case for AI agents is being able to extract out items from a diverse set of complex documents in a repeatable manner - whether it’s legal contracts, invoices, financial statements, passports, and more. In this case, 🚫 prompting ChatGPT/Claude is too error-prone/not systematic. 💡We’ve created a plug-and-play MCP server around our core document extraction capabilities within LlamaCloud. 1. Build an extraction agent in LlamaCloud with a standardized schema. 2. Plug it as an MCP tool within ChatGPT/Claude. 3. For your next extraction job, get the outputs in a standardized format through the tool! Huge shoutout to Tuana and Logan Markewich for working on this - this is thanks to the latest updates to our llamacloud-mcp demo repo, which supports both extraction and retrieval as MCP tools. Repo: Sign up:

Jerry Liu

57,827 Aufrufe • vor 1 Jahr

Take a first look at the Stake Engine back office! From signing up and creating your team to uploading your games directly to Stake. It all starts here. Coming soon: - Building with our tools - Payments & analytics insights - And much more! Stay tuned.

Take a first look at the Stake Engine back office! From signing up and creating your team to uploading your games directly to Stake. It all starts here. Coming soon: - Building with our tools - Payments & analytics insights - And much more! Stay tuned.

Stake Engine RGS

87,966 Aufrufe • vor 1 Jahr

Elon Musk: Grokipedia will succeed, I think it's fundamentally a superior product to Wikipedia. It is a better source of information and we haven't even added images and video yet "When you read a Grokipedia article about a subject or a person that you know a lot about, and you see, 'Wow, this is way better than Wikipedia. It's more comprehensive, it's way more accurate, it’s neutral instead of biased'....then you're gonna forward those links around and say, 'This is actually the better source' Grokipedia will succeed, I think very well, because it is fundamentally a superior product to Wikipedia. It is a better source of information. And we haven't even added images and video yet And then we used that latest version of Grok to create Grokipedia, which I think is it's not just, I think, more neutral and more accurate than Wikipedia, but actually has a lot more information than a Wikipedia page"

Elon Musk: Grokipedia will succeed, I think it's fundamentally a superior product to Wikipedia. It is a better source of information and we haven't even added images and video yet "When you read a Grokipedia article about a subject or a person that you know a lot about, and you see, 'Wow, this is way better than Wikipedia. It's more comprehensive, it's way more accurate, it’s neutral instead of biased'....then you're gonna forward those links around and say, 'This is actually the better source' Grokipedia will succeed, I think very well, because it is fundamentally a superior product to Wikipedia. It is a better source of information. And we haven't even added images and video yet And then we used that latest version of Grok to create Grokipedia, which I think is it's not just, I think, more neutral and more accurate than Wikipedia, but actually has a lot more information than a Wikipedia page"

X Freeze

109,095 Aufrufe • vor 9 Monaten

today, we're launching Light Parse Extend can now parse documents up to 70% cheaper than other solutions - light parse starts at $0.00625 per page, and scales down significantly with volume discounts - it's SOTA on Databricks OfficeQA Pro (a 3rd party agent benchmark) among all non-agentic parsing providers our ML/AI team trained smaller and more efficient layout, table, and form vision models to enable significant cost reduction without impacting accuracy on many types of documents live today for all users!

today, we're launching Light Parse Extend can now parse documents up to 70% cheaper than other solutions - light parse starts at $0.00625 per page, and scales down significantly with volume discounts - it's SOTA on Databricks OfficeQA Pro (a 3rd party agent benchmark) among all non-agentic parsing providers our ML/AI team trained smaller and more efficient layout, table, and form vision models to enable significant cost reduction without impacting accuracy on many types of documents live today for all users!

Kushal Byatnal

48,543 Aufrufe • vor 16 Tagen

Why you should NOT use Supabase: 1. SDK Trap - not TypeScript by default, type-gen is bad, easy to make mistakes 2. Security with the SDK is terrible - need to set up a lot of things to have something work 3. Vendor lock-in with the Auth, don't have your keys 4. Pricing way more expensive than a $8/month db + better-auth 5. Better alternative for all tools they provide

Why you should NOT use Supabase: 1. SDK Trap - not TypeScript by default, type-gen is bad, easy to make mistakes 2. Security with the SDK is terrible - need to set up a lot of things to have something work 3. Vendor lock-in with the Auth, don't have your keys 4. Pricing way more expensive than a $8/month db + better-auth 5. Better alternative for all tools they provide

Melvyn • Builder

26,772 Aufrufe • vor 5 Monaten

AI agents move fast. So should your search foundation. Introducing the next generation of Amazon OpenSearch Serverless. We rebuilt OpenSearch Serverless from the ground up for agentic AI and unpredictable workloads: - Create resources in seconds - Autoscale up to 20x faster than before - Usage-based pricing with decoupled storage and compute - No idle costs with scale to zero - Up to 60% cost savings vs. provisioned clusters for peak capacity Learn more.

AI agents move fast. So should your search foundation. Introducing the next generation of Amazon OpenSearch Serverless. We rebuilt OpenSearch Serverless from the ground up for agentic AI and unpredictable workloads: - Create resources in seconds - Autoscale up to 20x faster than before - Usage-based pricing with decoupled storage and compute - No idle costs with scale to zero - Up to 60% cost savings vs. provisioned clusters for peak capacity Learn more.

Amazon Web Services

12,135 Aufrufe • vor 3 Monaten

Still currently a Work in Progress, but the rain in our PvE Only, Extraction Shooter, is shaping up pretty nicely... We also know how big optimization is, and as we are now upgrading to Unreal Engine 5.6, this opens up more opportunities for that. Stay tuned for our Dev Update #8 coming soon! Wishlist/Buy our game now.

Still currently a Work in Progress, but the rain in our PvE Only, Extraction Shooter, is shaping up pretty nicely... We also know how big optimization is, and as we are now upgrading to Unreal Engine 5.6, this opens up more opportunities for that. Stay tuned for our Dev Update #8 coming soon! Wishlist/Buy our game now.

Incursion Red River

63,683 Aufrufe • vor 10 Monaten

📰 For the first time ever --> Gym Class VR now support full-court games in NBA team courts with v1.0.672. Hop on your Meta Quest and experience it now! Details in video and below .... - Squad up with your team. Set your NBA team court settings to 2v2 or 3v3 full-court to get started. 4v4 and 5v5 is available in beta environments and will be ramping up soon. The latter will start with Meta Quest 3 players and expand from there. - Fast paced fast breaks. Run down the court to score or chase down the play. More locomotion controls unlocks a whole new style of play. - High IQ inbounds. Read the defense to run your play. Added rules like 8 second violation, backcourt, and more. This release still has a few known bugs that we are working on in parallel. Expect a ton of updates coming to full-court soon and in the future. Over coming weeks, we plan to open up the mode to Pro courts and then Public as well.

📰 For the first time ever --> Gym Class VR now support full-court games in NBA team courts with v1.0.672. Hop on your Meta Quest and experience it now! Details in video and below .... - Squad up with your team. Set your NBA team court settings to 2v2 or 3v3 full-court to get started. 4v4 and 5v5 is available in beta environments and will be ramping up soon. The latter will start with Meta Quest 3 players and expand from there. - Fast paced fast breaks. Run down the court to score or chase down the play. More locomotion controls unlocks a whole new style of play. - High IQ inbounds. Read the defense to run your play. Added rules like 8 second violation, backcourt, and more. This release still has a few known bugs that we are working on in parallel. Expect a ton of updates coming to full-court soon and in the future. Over coming weeks, we plan to open up the mode to Pro courts and then Public as well.

Gym Class

13,820 Aufrufe • vor 2 Jahren

🚨 SAM ALTMAN: "A REDUCTION IN COST OF ABOUT A 1000X" 🤯 Sam Altman revealed the cost to solve a hard reasoning problem has plummeted by 1,000x since the release of o1 just 16 months ago (comparing o1 to the new 5.4 model). How is this possible? Altman credits two main factors: 1️⃣ We're Still in the Early Days: “We are so new to this paradigm that there is massive room for improvement. We're still doing things in "dumb ways" and getting better fast.” 2️⃣ Human Ingenuity is Underrated: “It's not just the models getting better. It's kernel engineers, power engineers, and data center designers finding brilliant ways to optimize every step of the process.” The compounding effect of better models + better human engineering is driving an exponential decrease in the cost of AI reasoning.

🚨 SAM ALTMAN: "A REDUCTION IN COST OF ABOUT A 1000X" 🤯 Sam Altman revealed the cost to solve a hard reasoning problem has plummeted by 1,000x since the release of o1 just 16 months ago (comparing o1 to the new 5.4 model). How is this possible? Altman credits two main factors: 1️⃣ We're Still in the Early Days: “We are so new to this paradigm that there is massive room for improvement. We're still doing things in "dumb ways" and getting better fast.” 2️⃣ Human Ingenuity is Underrated: “It's not just the models getting better. It's kernel engineers, power engineers, and data center designers finding brilliant ways to optimize every step of the process.” The compounding effect of better models + better human engineering is driving an exponential decrease in the cost of AI reasoning.

Chris

283,126 Aufrufe • vor 6 Monaten

ChatGPT o1 is the most "intelligent" AI model and it's not even close! Full o1 generates thinking steps ~50 faster than preview. It's more accurate, reliable, and got better on harder tasks that require advanced reasoning and knowledge. I ran a few tests on it already. Here are my observations: Full video with examples & explanations: Strengths - impressive at math, code, and knowledge-intensive tasks. Weakness - it only failed on a cross-word puzzle but I think it might be solvable when a web search becomes available. In the end, while very efficient with complex knowledge use, it's still constrained by data it's trained on. Speed - the thinking steps are generated a lot faster! Not a fair comparison with the open alternatives but I think this improves the overall user experience. "Knowledgeable and highly intelligent" - as mentioned in the demo by OpenAI researchers, o1 is great at dealing with ambiguity and filling in knowledge gaps. I was impressed by how it implemented an agentic solution (with lots of details) from a basic diagram of architecture (with minimal details). Check out the sample video. Better Task Coverage - Due to the speed and the ability to make sense of instructions and intent (i.e., know when to response fast and when to "think" deeply) much better, it feels like it might be more useful for a broader range of tasks. Image understanding - the image understanding capability is mysterious (often leads to faster responses but no thinking) but impressive. More experiments and notes soon. Stay tuned!

ChatGPT o1 is the most "intelligent" AI model and it's not even close! Full o1 generates thinking steps ~50 faster than preview. It's more accurate, reliable, and got better on harder tasks that require advanced reasoning and knowledge. I ran a few tests on it already. Here are my observations: Full video with examples & explanations: Strengths - impressive at math, code, and knowledge-intensive tasks. Weakness - it only failed on a cross-word puzzle but I think it might be solvable when a web search becomes available. In the end, while very efficient with complex knowledge use, it's still constrained by data it's trained on. Speed - the thinking steps are generated a lot faster! Not a fair comparison with the open alternatives but I think this improves the overall user experience. "Knowledgeable and highly intelligent" - as mentioned in the demo by OpenAI researchers, o1 is great at dealing with ambiguity and filling in knowledge gaps. I was impressed by how it implemented an agentic solution (with lots of details) from a basic diagram of architecture (with minimal details). Check out the sample video. Better Task Coverage - Due to the speed and the ability to make sense of instructions and intent (i.e., know when to response fast and when to "think" deeply) much better, it feels like it might be more useful for a broader range of tasks. Image understanding - the image understanding capability is mysterious (often leads to faster responses but no thinking) but impressive. More experiments and notes soon. Stay tuned!

elvis

102,643 Aufrufe • vor 1 Jahr

Armando Duka: "We should have one national team and one country. 🇦🇱🇽🇰 But that's not up to me." Interviewer: "But maybe we'll have twice the joy if both teams win." Duka: "No, I think it's better together, we have more potential, we are bigger, it's better in every way, not just from a football perspective."

Armando Duka: "We should have one national team and one country. 🇦🇱🇽🇰 But that's not up to me." Interviewer: "But maybe we'll have twice the joy if both teams win." Duka: "No, I think it's better together, we have more potential, we are bigger, it's better in every way, not just from a football perspective."

Albanian Football Scout 🇦🇱

25,248 Aufrufe • vor 8 Monaten

In this video, Jacob had an issue standing up, flipping a little bit, losing his posture, and that club is releasing early. Now, in this after video, we're going to see him opening up the hips much better, we see him release the club, much more in front. And now he's got a lot more consistency in his swing, a lot better contact when that's happening. ⬇️See how the B29 Blue Brick can help you solve this common problem in as little as one range session ⬇️

In this video, Jacob had an issue standing up, flipping a little bit, losing his posture, and that club is releasing early. Now, in this after video, we're going to see him opening up the hips much better, we see him release the club, much more in front. And now he's got a lot more consistency in his swing, a lot better contact when that's happening. ⬇️See how the B29 Blue Brick can help you solve this common problem in as little as one range session ⬇️

Top Speed Golf

33,639 Aufrufe • vor 4 Monaten

Tyrese Maxey, on not making the All-Star team: "I'm happy, actually. I have a lot of connection with the people that made it...Everybody is deserving. That doesn't upset me at all. I'm more focused on trying to help this team win and get better."

Tyrese Maxey, on not making the All-Star team: "I'm happy, actually. I have a lot of connection with the people that made it...Everybody is deserving. That doesn't upset me at all. I'm more focused on trying to help this team win and get better."

Derek Bodner

1,051,334 Aufrufe • vor 1 Jahr

🛑 Sneak Peek Alert! 🛑 Hey, LETSTOP fam! Are you ready to revolutionize your driving experience? Buckle up, because we're about to give you a sneak peek into our upcoming app! 📱 Get ready to explore a cutting-edge safe driving app designed to make every journey safer and more rewarding. 🏆 LETSTOP is not just an app—it's a movement towards a safer, smarter, and more rewarding driving culture. Stay tuned for more updates as we gear up for the big launch together! Keep an eye out for more updates coming soon! #LETSTOP

🛑 Sneak Peek Alert! 🛑 Hey, LETSTOP fam! Are you ready to revolutionize your driving experience? Buckle up, because we're about to give you a sneak peek into our upcoming app! 📱 Get ready to explore a cutting-edge safe driving app designed to make every journey safer and more rewarding. 🏆 LETSTOP is not just an app—it's a movement towards a safer, smarter, and more rewarding driving culture. Stay tuned for more updates as we gear up for the big launch together! Keep an eye out for more updates coming soon! #LETSTOP

LETSTOP

42,192 Aufrufe • vor 2 Jahren

.buffalu on Solana: "There's the SIMD and protocol improvements around inflation and burning, there's tokenized equities, and a lot of stuff we're going to bring via JTX to get more people trading on Solana" "The best teams win by iterating the fastest, and no one is iterating faster than Solana. Not only the teams that exist today but the new teams coming up too" "You can look at the current state of crypto and be bearish, but it's hard to do that if you look at the state of iteration on Solana"

.buffalu on Solana: "There's the SIMD and protocol improvements around inflation and burning, there's tokenized equities, and a lot of stuff we're going to bring via JTX to get more people trading on Solana" "The best teams win by iterating the fastest, and no one is iterating faster than Solana. Not only the teams that exist today but the new teams coming up too" "You can look at the current state of crypto and be bearish, but it's hard to do that if you look at the state of iteration on Solana"

Solana

127,243 Aufrufe • vor 1 Monat

After 30 years of disinvestment in higher education, the cost to attend our colleges and universities has skyrocketed. We needed to do better. That’s why Pennsylvania’s Secretary of Education and his team went to work — and presented a blueprint to make higher education more affordable and accessible for all Pennsylvanians. And after Democrats and Republicans came together, we invested more in our HBCUs, community colleges, and Pennsylvania's State System of Higher Education institutions — all while investing more money in new scholarships to help bring down the cost of higher education.

After 30 years of disinvestment in higher education, the cost to attend our colleges and universities has skyrocketed. We needed to do better. That’s why Pennsylvania’s Secretary of Education and his team went to work — and presented a blueprint to make higher education more affordable and accessible for all Pennsylvanians. And after Democrats and Republicans came together, we invested more in our HBCUs, community colleges, and Pennsylvania's State System of Higher Education institutions — all while investing more money in new scholarships to help bring down the cost of higher education.

Governor Josh Shapiro

35,560 Aufrufe • vor 2 Jahren

Big moment for Postgres! AI coding tools have been surprisingly bad at writing Postgres code. Not because the models are dumb, but because of how they learned SQL in the first place. LLMs are trained on the internet, which is full of outdated Stack Overflow answers and quick-fix tutorials. So when you ask an AI to generate a schema, it gives you something that technically runs but misses decades of Postgres evolution, like: - No GENERATED ALWAYS AS IDENTITY (added in PG10) - No expression or partial indexes - No NULLS NOT DISTINCT (PG15) - Missing CHECK constraints and proper foreign keys - Generic naming that tells you nothing But this is actually a solvable problem. You can teach AI tools to write better Postgres by giving them access to the right documentation at inference time. This exact solution is actually implemented in the newly released pg-aiguide by Tiger Data - Creators of TimescaleDB, which is an open-source MCP server that provides coding tools access to 35 years of Postgres expertise. In a gist, the MCP server enables: - Semantic search over the official PostgreSQL manual (version-aware, so it knows PG14 vs PG17 differences) - Curated skills with opinionated best practices for schema design, indexing, and constraints. I ran an experiment with Claude Code to see how well this works, and worked with the team to put this together. Prompt: "Generate a schema for an e-commerce site twice, one with the MCP server disabled, one with it enabled. Finally, run an assessment to compare the generated schemas." The run with the MCP server led to: - 420% more indexes (including partial and expression indexes) - 235% more constraints - 60% more tables (proper normalization) - 11 automation functions and triggers - Modern PG17 patterns throughout The MCP-assisted schema had proper data integrity, performance optimizations baked in, and followed naming conventions that actually make sense in production. pg-aiguide works with Claude Code, Cursor, VS Code, and any MCP-compatible tool. It's free and fully open source. I have shared the repo in the replies!

Big moment for Postgres! AI coding tools have been surprisingly bad at writing Postgres code. Not because the models are dumb, but because of how they learned SQL in the first place. LLMs are trained on the internet, which is full of outdated Stack Overflow answers and quick-fix tutorials. So when you ask an AI to generate a schema, it gives you something that technically runs but misses decades of Postgres evolution, like: - No GENERATED ALWAYS AS IDENTITY (added in PG10) - No expression or partial indexes - No NULLS NOT DISTINCT (PG15) - Missing CHECK constraints and proper foreign keys - Generic naming that tells you nothing But this is actually a solvable problem. You can teach AI tools to write better Postgres by giving them access to the right documentation at inference time. This exact solution is actually implemented in the newly released pg-aiguide by Tiger Data - Creators of TimescaleDB, which is an open-source MCP server that provides coding tools access to 35 years of Postgres expertise. In a gist, the MCP server enables: - Semantic search over the official PostgreSQL manual (version-aware, so it knows PG14 vs PG17 differences) - Curated skills with opinionated best practices for schema design, indexing, and constraints. I ran an experiment with Claude Code to see how well this works, and worked with the team to put this together. Prompt: "Generate a schema for an e-commerce site twice, one with the MCP server disabled, one with it enabled. Finally, run an assessment to compare the generated schemas." The run with the MCP server led to: - 420% more indexes (including partial and expression indexes) - 235% more constraints - 60% more tables (proper normalization) - 11 automation functions and triggers - Modern PG17 patterns throughout The MCP-assisted schema had proper data integrity, performance optimizations baked in, and followed naming conventions that actually make sense in production. pg-aiguide works with Claude Code, Cursor, VS Code, and any MCP-compatible tool. It's free and fully open source. I have shared the repo in the replies!

Avi Chawla

187,076 Aufrufe • vor 8 Monaten