extrakt.AI

extrakt.AI

Software Development

Palo Alto, California 498 followers

Building AI Agents that Automate Supply Chain Operations in a Day

About us

extrakt.AI employs AI agents that automate Supply Chain Operations in a day. Our AI agents extract facts from supply chain documentation and communication, such as contracts, delivery notes, inquiries, and orders, directly from emails. They resolve ambiguous or incorrect data based on context, sending the information to ERP, DMS, or TMS systems without the need for manual data entry. We use GenAI to make process templates super simple to create, starting with just a basic outline

Website
https://www.extrakt.ai/
Industry
Software Development
Company size
2-10 employees
Headquarters
Palo Alto, California
Type
Privately Held
Founded
2023
Specialties
ai, enterprise software, B2B, logistics, supply chain, and AI Agents

Products

Locations

Employees at extrakt.AI

Updates

  • 𝗔𝗜 𝘁𝗵𝗿𝗶𝘃𝗲𝘀 𝗼𝗻 𝗰𝗼𝗹𝗹𝗮𝗯𝗼𝗿𝗮𝘁𝗶𝗼𝗻, 𝗮𝗻𝗱 𝗼𝘂𝗿 𝗳𝗼𝘂𝗻𝗱𝗶𝗻𝗴 𝗔𝗜 𝗘𝗻𝗴𝗶𝗻𝗲𝗲𝗿 Jan Rupnik 𝗶𝘀 𝗹𝗲𝗮𝗱𝗶𝗻𝗴 𝘁𝗵𝗲 𝗰𝗵𝗮𝗿𝗴𝗲 🚀 For over 20 years, Jan has been deeply immersed in Artificial Intelligence research and development, driving innovation in AI for supply chains. But his passion doesn’t stop there! Jan is dedicated to expanding AI know-how within the broader tech community. Every 2 weeks, he leads the 𝗟𝗟𝗠 𝗗𝗼𝗷𝗼 - a hybrid gathering of engineers and AI enthusiasts exploring cutting-edge topics like transformers, efficient inference, synthetic data generation, and more. The goal is to share knowledge, tackle challenges, and push the boundaries of what’s possible with AI. 📅 Want to join the conversation? DM Jan to become part of the 𝗟𝗟𝗠 𝗗𝗼𝗷𝗼 𝗖𝗼𝗺𝗺𝘂𝗻𝗶𝘁𝘆, or share this post with someone who might be interested in these dynamic discussions. Because the future of AI isn’t a one-man show or a lone-wolf mission - it’s a pack effort driven by shared vision. ⚡ 

    🚀 Exciting update from today's LLM Dojo meeting 🚀 We had a great session today where we looked at what happened in generative AI last month (tech/engineering oriented) - 🤯 🔍 Today's focus: a cool and fast growing library: browser_use (https://lnkd.in/dARw3KC2) 💡 The nice thing about it is that browser automations based on LLMs are a lot cheaper than the alternatives I know since the approach can work with a textual representation of interactive parts of the DOM 👏 ✨ 🚨 Want to join? Anyone is interested in participating at our meetings, DM me and I'll add you to our group. The meetings are hybrid (every second Friday, 9:00 AM Ljubljana time). Topics we’ll cover include: • Transformers • Efficient inference • Reasoning and problem-solving capabilities • Synthetic data generation • Test-time compute strategies • Emerging LLM architectures • Automatic prompt engineering • Agentic frameworks • Interpretability and control • Tools and platforms for LLM development • Building applications on top of LLMs • Interesting papers and current research • And much more…

    GitHub - browser-use/browser-use: Make websites accessible for AI agents

    GitHub - browser-use/browser-use: Make websites accessible for AI agents

    github.com

  • We’re back! 🎉 We had a blast in 2024, and apparently, so did our clothes—they've mysteriously and collectively shrunk over the holidays! 👖🧐 Fear not, we're squeezing ourselves back into the office and gearing up to tackle all the exciting challenges ✨ 2025 ✨ throws our way. Stay tuned for some thrilling updates coming your way soon! 😉 

    • No alternative text description for this image
  • 🎉 Less hassle, more flow in 2025! As we raise a toast to 2024, we want to thank you for trusting us to transform your supply chain with AI-powered automation. Here’s to an efficient and rewarding year ahead! 🚀

    • No alternative text description for this image
  • Can GenAI leverage never-ending email exchanges? Can it make sense of all the scattered information and automate data processing for the supply chain? ⚙️👷♂️ Welcome to 𝗔𝗜 𝗠𝘆𝘁𝗵𝗯𝘂𝘀𝘁𝗶𝗻𝗴 𝗖𝗵𝗮𝗹𝗹𝗲𝗻𝗴𝗲 #𝟭𝟰 🎯 Processing supply chain data from email threads is messy and inefficient, with delivery updates often buried in lengthy conversations. Without GenAI’s ability to organize and reason through this data, tracking accurate delivery statuses become a hurdle. So, how does GenAI tackle this challenge? Swipe to the end to find out the answer 👆 #AI #ArtificialIntelligence #TechTrends #Innovation #MachineLearning #AITechnology #DataScience #TechCommunity #AIInsights #futureofwork

  • Automation can save you a ton of effort, but it doesn’t spare you from giving it a split-second attention. 𝗟𝗼𝗻𝗴 𝘀𝘁𝗼𝗿𝘆: While prepping for our webinar, Tyler and Veronika went full geek mode testing their complex Power Automate flow:     📄 PDF Invoice arrives in email  ➡️ Is forwarded to SharePoint  🏁 Document and data land in the SharePoint document library All perfect in theory: ✅ Invoice document sent  ✅ Flow set up correctly  ✅ Invoice data accurate Still, in testing, the flow was failing, and they couldn’t tell why. Turns out Veronika sent a JPG instead of a PDF. Yup, the almighty automation was no match for one tiny little photo 🤦♀ 𝗟𝗼𝗻𝗴 𝘀𝘁𝗼𝗿𝘆 𝘀𝗵𝗼𝗿𝘁:   𝗘𝘃𝗲𝗻 𝘁𝗵𝗲 𝗳𝗮𝗻𝗰𝗶𝗲𝘀𝘁 𝗣𝗼𝘄𝗲𝗿 𝗔𝘂𝘁𝗼𝗺𝗮𝘁𝗲 𝗳𝗹𝗼𝘄 𝗰𝗮𝗻’𝘁 𝘀𝗮𝘃𝗲 𝘆𝗼𝘂 𝗳𝗿𝗼𝗺 𝗮 𝗴𝗼𝗼𝗱 𝗼𝗹𝗱-𝗳𝗮𝘀𝗵𝗶𝗼𝗻𝗲𝗱 “𝗼𝗼𝗽𝘀.” So, while automation rocks, remember sometimes it’s only as human-proof as its users 🚀 Want to see how it all came together in the end? Watch the webinar where everything worked like a charm! 👇  https://lnkd.in/dgTKXkfr

    • No alternative text description for this image
  • GenAI is shifting from experimentation to a strategic stage. Enterprises now prioritize measurable ROI, industry-specific customization, and tools that truly perform over mere clout. More insights on the state of GenAI from our co-founder Tomaz Suklje, particularly the main reasons for accelerated adoption across enterprises 👇

    View profile for Tomaz Suklje, graphic

    Co-founder and Co-CEO @extrakt.AI | @senzemo | Supply Chain AI Agents

    In the past, 𝗚𝗲𝗻𝗔𝗜 𝗮𝗱𝗼𝗽𝘁𝗶𝗼𝗻 𝘂𝘀𝗲𝗱 𝘁𝗼 𝗯𝗲 𝗹𝗼𝗼𝘀𝗲 𝗳𝗿𝗼𝗺 𝗯𝘂𝘀𝗶𝗻𝗲𝘀𝘀 𝗼𝗯𝗷𝗲𝗰𝘁𝗶𝘃𝗲𝘀. The focus was more on novelty rather than practicality. ⏮ 𝗡𝗼𝘄, 𝘄𝗲’𝗿𝗲 𝘀𝗲𝗲𝗶𝗻𝗴 𝗚𝗲𝗻𝗔𝗜 𝘀𝗵𝗶𝗳𝘁 𝗳𝗿𝗼𝗺 𝗲𝘅𝗽𝗲𝗿𝗶𝗺𝗲𝗻𝘁𝗮𝘁𝗶𝗼𝗻 𝘁𝗼 𝗰𝗼𝗿𝗲 𝘀𝘁𝗿𝗮𝘁𝗲𝗴𝘆. 🔄 𝗔𝗻𝗱 𝗲𝗻𝘁𝗲𝗿𝗽𝗿𝗶𝘀𝗲𝘀 𝗮𝗿𝗲 𝗺𝗼𝘀𝘁 𝗼𝗳 𝗮𝗹𝗹 𝗮𝘀𝗸𝗶𝗻𝗴 𝗳𝗼𝗿 𝘁𝗮𝗻𝗴𝗶𝗯𝗹𝗲 𝗿𝗲𝘀𝘂𝗹𝘁𝘀:    1. Immediate value beats the fear of the unknown.   Only 1% decision-makers choose based on familiarity.    2. As long as GenAI delivers immediate, measurable value and personalization, people no longer care how popular the tool is when they make a decision: - The need for easily quantifiable ROI counts for 30%   - Capacity for industry/organization customization weighs 26%  - Good reputation counts only as much as 1%     3. Enterprises need to see that GenAI actually “works on their machines”. The fastest way to see it happen is bottom-up, case by case, where bandages no longer stick: - 28% of cases are about enterprise research & retrieval  - 27% focus on data extraction & transformation  - 10% on workflow automation    Get more data-backed insights from Menlo Ventures, in their State of Generative AI in Enterprise Report: 💎 https://lnkd.in/dYWNgYDV Or better yet, let’s talk about this. 

  • 2️⃣ More Days to Go! 🚀 Join us and Tyler Wasson as we dive into a demo on quick & dirty automation with Power Automate and SharePoint.

    View organization page for extrakt.AI, graphic

    498 followers

    Invoices everywhere, manual data entries driving you mad? Let AI bring order to chaos and help you get the most out of your SharePoint. Give yourself 30 minutes and join our webinar next week. 𝗣𝗼𝘄𝗲𝗿 𝗔𝘂𝘁𝗼𝗺𝗮𝘁𝗲 𝗘𝘅𝗽𝗲𝗿𝘁 Tyler Wasson & 𝗼𝘂𝗿 𝘁𝗲𝗮𝗺 will show you how to put AI agents to work and automate your invoice processing. It’s all about saving you time and sparing you the hassle of mistakes. You will learn how to: 🔵 Automatically extract and update invoice data 🔵 Eliminate repetitive tasks and manual data errors with AI 🔵 Maximize investment in your SharePoint document management system Hit 𝗔𝘁𝘁𝗲𝗻𝗱 and join us to make manual entry a thing of the past!

    This content isn’t available here

    Access this content and more in the LinkedIn app

  • Meet our partner and webinar guest! 👋 With over 2️⃣0️⃣ years of experience across retail, manufacturing, and IT, Tyler Wasson is a seasoned expert in the Microsoft Power Platform, including Power BI, Power Automate, and Power Apps. We’re excited to have Tyler as our webinar guest on 𝗗𝗲𝗰𝗲𝗺𝗯𝗲𝗿 𝟱𝘁𝗵, as he’ll be shedding light on how to 𝗮𝘂𝘁𝗼𝗺𝗮𝘁𝗶𝗰𝗮𝗹𝗹𝘆 𝗽𝗿𝗼𝗰𝗲𝘀𝘀 𝗶𝗻𝘃𝗼𝗶𝗰𝗲𝘀 𝗶𝗻 𝗦𝗵𝗮𝗿𝗲𝗣𝗼𝗶𝗻𝘁 in a fast, easy-to-understand way. If invoice processing is a hassle for you, we hope you can join us! 👨💻 Register for the webinar here: https://lnkd.in/dcM-kgRV

    • No alternative text description for this image

Similar pages

Funding

extrakt.AI 1 total round

Last Round

Pre seed
See more info on crunchbase