Reworkd uses LLM to extract web data at scale. Our platform generates and repairs Playwright scraping code for thousands of websites automatically. No more maintaining scrapers—just provide feedback on issues and our AI fixes them instantly.
We’re super excited today to announce that we're launching Reworkd: your AI scraping co-pilot.
Our company journey first started with AgentGPT which hit 1M users and 30k GitHub stars in just a few months. From there we went through YC and spent many, many months talking to users and iterating in the LLM agent space. Many micro pivots later, we eventually discovered that many companies were struggling with the same problem: managing the engineering burden of building and maintaining scrapers across hundreds of sites.
So we build the platform specifically for companies that need to regularly scrape the same 10s if not 100s of websites to power their products.
Here are some of the things that make us different:
Our platform automatically generates Playwright code to extract website data into your custom schema.
You have full autonomy over this code: you can provide additional guidance in chat and even update the code in our built-in IDE
We handle all of the proxy and antibot measures for you.
You can schedule scrapers to run at your desired frequency, choose between full or incremental scraping, and deduplicate data based on a primary key.
Easily ingested all of your scraped data via an API
We've been working with power users who scrape 100s of sites, and they've helped us refine the platform over the past year. Use cases we're seeing include:
E-commerce product data for AI shopping solutions or price monitoring
Government sites for bids, rules, and regulations
Company career pages for job postings
Real estate listings from brokers and MLS systems
If you're scraping 10+ sites regularly, we’d love your feedback—try our platform and let us know what you think. Product Hunters: enjoy 50% off your first month!
Looking forward to your thoughts on how we can make this better 🫡
Report
@srijan__subedi This one hits a real pain point. Scraping at scale is messy, brittle, and always more work than expected. Love how your team gives organisations both control and automation without drowning in infra or anti-bot work. The AgentGPT roots + real use cases add solid credibility.
Honestly, a sharp 60-sec visual showing “before Reworkd vs. after” could really help new users get the magic faster.
Big props to the team — excited to see this scale!
Based on community feedback, we're hoping to also open source our browser fleet orchestrator and arbitrary code execuction service (which uses firecracker under the hood).
Happy to answer any and all technical questions on our stack! Really appreciate all the support! 🙂
I'm Asim- one of the co-founders @Reworkd . We've spend many hours tinkering, fine tuning, and bug fixing (quite literally till the last minute of launch) which is why we're all super excited to finally get our platform in your hands.
We'll be around to answer any questions and take any feed back good or bad. Thank you for the support ❤️
Report
I have been following you guys for years and super excited by what you have built.
It's been a while I have been following you guys and I am super excited to see @Reworkd growth. Super proud of you and your team. Wish you a good luck.
We’re super excited today to announce that we're launching Reworkd: your AI scraping co-pilot.
Our company journey first started with AgentGPT which hit 1M users and 30k GitHub stars in just a few months. From there we went through YC and spent many, many months talking to users and iterating in the LLM agent space. Many micro pivots later, we eventually discovered that many companies were struggling with the same problem: managing the engineering burden of building and maintaining scrapers across hundreds of sites.
So we build the platform specifically for companies that need to regularly scrape the same 10s if not 100s of websites to power their products.
Here are some of the things that make us different:
Our platform automatically generates Playwright code to extract website data into your custom schema.
You have full autonomy over this code: you can provide additional guidance in chat and even update the code in our built-in IDE
We handle all of the proxy and antibot measures for you.
You can schedule scrapers to run at your desired frequency, choose between full or incremental scraping, and deduplicate data based on a primary key.
Easily ingested all of your scraped data via an API
We've been working with power users who scrape 100s of sites, and they've helped us refine the platform over the past year. Use cases we're seeing include:
E-commerce product data for AI shopping solutions or price monitoring
Government sites for bids, rules, and regulations
Company career pages for job postings
Real estate listings from brokers and MLS systems
If you're scraping 10+ sites regularly, we’d love your feedback—try our platform and let us know what you think. Product Hunters: enjoy 50% off your first month!
Looking forward to your thoughts on how we can make this better 🫡
Reworkd is highly praised for its efficient web data extraction capabilities, with users appreciating its ease of use and effectiveness. A notable endorsement comes from the makers of
, who commend Reworkd for its excellent support and personalized communication, preferring it over larger companies. Users consistently highlight the platform's ability to handle complex scraping tasks effortlessly, making it a reliable choice for those needing scalable data extraction solutions.
Would love to give a major shout out to our scraping partners, Reworkd. They launched recently on Product Hunt. They've been excellent every step as we've built. We considered using bigger scraping companies like Zyte, but we didn't want to use a large company. I get to talk to the Reworkd founders every day in Slack.
Astrid: Personal Shopping Agent
Reworkd
Hey Product Hunt! 👋
We’re super excited today to announce that we're launching Reworkd: your AI scraping co-pilot.
Our company journey first started with AgentGPT which hit 1M users and 30k GitHub stars in just a few months. From there we went through YC and spent many, many months talking to users and iterating in the LLM agent space. Many micro pivots later, we eventually discovered that many companies were struggling with the same problem: managing the engineering burden of building and maintaining scrapers across hundreds of sites.
So we build the platform specifically for companies that need to regularly scrape the same 10s if not 100s of websites to power their products.
Here are some of the things that make us different:
Our platform automatically generates Playwright code to extract website data into your custom schema.
You have full autonomy over this code: you can provide additional guidance in chat and even update the code in our built-in IDE
We handle all of the proxy and antibot measures for you.
You can schedule scrapers to run at your desired frequency, choose between full or incremental scraping, and deduplicate data based on a primary key.
Easily ingested all of your scraped data via an API
We've been working with power users who scrape 100s of sites, and they've helped us refine the platform over the past year. Use cases we're seeing include:
E-commerce product data for AI shopping solutions or price monitoring
Government sites for bids, rules, and regulations
Company career pages for job postings
Real estate listings from brokers and MLS systems
If you're scraping 10+ sites regularly, we’d love your feedback—try our platform and let us know what you think. Product Hunters: enjoy 50% off your first month!
Looking forward to your thoughts on how we can make this better 🫡
@srijan__subedi This one hits a real pain point. Scraping at scale is messy, brittle, and always more work than expected. Love how your team gives organisations both control and automation without drowning in infra or anti-bot work. The AgentGPT roots + real use cases add solid credibility.
Honestly, a sharp 60-sec visual showing “before Reworkd vs. after” could really help new users get the magic faster.
Big props to the team — excited to see this scale!
Reworkd
Hey Friends 👋
I am Adam, the CTO here @Reworkd. Our team is super passionate about OSS and we are actively working on open sourcing more parts of our stack. We've already open sourced our Agent Eval Framework, our Extraction SDK and a LLM Vision toolkit.
Based on community feedback, we're hoping to also open source our browser fleet orchestrator and arbitrary code execuction service (which uses firecracker under the hood).
Happy to answer any and all technical questions on our stack! Really appreciate all the support! 🙂
Reworkd
Hey everyone,
I'm Asim- one of the co-founders @Reworkd . We've spend many hours tinkering, fine tuning, and bug fixing (quite literally till the last minute of launch) which is why we're all super excited to finally get our platform in your hands.
We'll be around to answer any questions and take any feed back good or bad. Thank you for the support ❤️
I have been following you guys for years and super excited by what you have built.
Congratulations!
Reworkd
Appreciate all the support you've given us @chrishobbsttt !
Reworkd
@chrishobbsttt Thanks Chris!! Appreciate the support:)
Hey Srijan & Reworkd Team.
It's been a while I have been following you guys and I am super excited to see @Reworkd growth. Super proud of you and your team. Wish you a good luck.
Dibyesh
Reworkd
@dibyesh_giri Really appreciate for the kind words:))
Blogwald
I found this tool to be super helpful for data extraction. When I was working on a scraping project, this was byfar the easiest solution to use.
Reworkd
@amrutha____ Appreciate it!!
BITHUB
You’ve put in the work, and it shows. Wishing you all the success in the world! 🙌
Reworkd
Appreciate it Mohamed @mohamed_zakarya ! Super proud of the team