SELECT * FROM Everything is live now! Join leaders from AppsFlyer, Crunchbase and Nym explore the practicalities of working with unstructured data in the wild, and get a glimpse into how some of the best minds in the industry tackle this extraordinary challenge. Watch the event to learn: ✅ What are the avenues you can take to unlock your unstructured data ✅ What an unstructured data stack looks like in practice ✅ What are the major pitfalls when trying to utilize freeform text data ✅ The realities of incorporating unstructured data into your structured dashboards ✅ What to think about when creating your own unstructured data strategy
Flexor
Data Infrastructure and Analytics
Tel-Aviv, Tel-Aviv District 794 followers
Structuring the Unstructured
About us
Flexor enables data practitioners to harness the power of LLMs on textual data. Flexor manages all your data preparation needs—extraction, cleansing, structuring, governance, and transformation—empowering data teams to extract valuable insights, gain a competitive edge, and drive innovation. Seamlessly integrate high-quality, well-governed textual data into any BI, ML, or GenAI application. Built for data teams aiming for unmatched accuracy and scalability in their data utilization.
- Website
-
https://flexor.ai
External link for Flexor
- Industry
- Data Infrastructure and Analytics
- Company size
- 11-50 employees
- Headquarters
- Tel-Aviv, Tel-Aviv District
- Type
- Privately Held
- Founded
- 2022
Locations
-
Primary
19 Yosef Karo
Tel-Aviv, Tel-Aviv District 6701423, IL
Employees at Flexor
-
Doron Veltzer
-
Ben Tytonovich
Founder @ Maple Capital | Partnering with technical founders from day-1 | Cyber, Data Infra, AI Delivery & Devops
-
Sharon Beilis
Head of Marketing at Flexor | Getting your unstructured data AI ready
-
Tom Granot
Fractional CMO | Technical Product Marketing | Helping Deeply Technical Startups with GTM | Generated 1000s of Qualified Technical Webinar Leads |…
Updates
-
It always warms our hearts to receive positive feedback after a great session with a client. It's even better when it comes from one of the hospitality industry's leading companies! 🙌 Challenges we tackled => Transforming heaps of customer reviews into meaningful deployable data. Next up => Developing account health scores to enhance their operations further! 📈 Want to see how the magic happens? Give us a shout at hello@flexor.ai
Flexor is making waves! It’s exciting to see our platform driving real value and sparking creativity among our customers. Saul Ron, our Head of Customer Success and Applied AI, showed me this Slack message from one of our customers earlier today. Tbh, I was excited. We're not an LLM. Flexor handles unstructured data end-to-end. Raw text → Deployable data. For any use case you can imagine. Try me. This is exactly why our platform makes frontier models look bad :)
-
📣 Attention CDOs - we're attending CDOIQ Boston! Will you be there? Let's connect and discuss how to prepare your unstructured textual data for generative AI. #CDOIQ #CDO ##CDAO #UnstructuredData #DataTransformation #DataLeaders #GenAI
-
Our team is traveling to the DataConnect Conference in Ohio tomorrow - Excited to connect with some of the greatest minds in data! Would you be around? Lets grab a coffee ☕ (or a beer 🍻) - we love talking about unstructured data and the practicalities of transforming your raw text into AI-ready data!
-
🗺️ Flexor is on the AI map! We’re excited to share that Flexor has been featured on the Bessemer Venture Partners Market Map for AI Infrastructure, in the "Pre-Processing and Pipelines" category! It’s an honor to be featured alongside industry leaders who are driving innovation in AI infrastructure ❣️ At Flexor, we transform unstructured text into a competitive edge. We simplify data processing, empowering companies to unlock insights and drive AI innovations from untapped textual data. A quote that caught our eyes: - - - “We believe a new generation of data ops giants will emerge fueled by a focus on AI workflows.” - - - Thank you Bessemer Venture Partners for this recognition, and a special thank you to Janelle Teng, Grace Ma Bhavik N., Mary D'Onofrio, Elliott Robinson, David Cowan for putting us on the map!
-
We're is heading to Columbus Ohio, for DataConnect Conference on July 11-12! If you're attending or in the area, drop us a message or comment below. Let's connect & explore how we can transform unstructured data together!
-
Taking LLM applications to production is challenging, BUT there are best practices to make the process efficient & quickly derive business insights. Want to learn how? Our very own CTO & Founder, Regev Ben Simon, will be speaking all about it this Thursday at the Google Cloud Summit. Discover how you can overcome these challenges and do it the Flexor way. If you’d like to catch up personally, DM us. See you there! #GoogleCloudSummit #DataStrategy #AIStrategy #DataLeaders" #LLM
-
Flexor reposted this
Data that engages business leaders? Gamers (hey #GDC2024), retailers, e-commerce data folks: you're gonna like this. The Game Developers Conference starts today, so I decided to highlight one of the world's most data-driven industries 🎮 Even if video games aren't your thing, I bet you're all about engaging your BUs and driving business impact, right? Imagine you could base analyses on what users say, and back it all with numbers. --- Say Hi to Palworld 🧸 ICYMI - Within 8 weeks of its release, the game Palworld already has 282,593 reviews on Steam. It became the most-played third-party release on Xbox Game Pass, and it reached a peak of nearly 3 million daily active players on the Xbox platform. TL;DR - (a) Palworld players want more pals! (b) They like it (way) more than Pokemon. Curious how we learned this? see below ⬇️ To understand what people think of the game and how can this hype can be turned into dollars, we analyzed all of those reviews using Flexor. To put things in perspective, an average Steam review contains 57 words. That means there are 16,107,801 words, or over 32,000 pages. Since we at Flexor enjoy working with a lot of text, let’s dig in. First and foremost - it turns out the vast majority of reviews are positive. Remember the plagiarism accusations? Apparently, most reviewers think Palworld IS BETTER than Pokemon! However, we thought it would be interesting to ask - what can we do to make it even better? We loaded all the reviews to our Snowflake instance and ran the following SQL query: SELECT FLEX(review_id, “what features does the reviewer request?”) FROM steam_reviews_dataset Reviewers raised lots of good ideas, like: - "There is a ton of potential for expansions. A water / diving update, building update, new maps and Pals." - "There are some things I would really hope they add in the future: some semblance of a story, and actual shiny pals." - "Looking forward to seeing new Pals added in the future." - "In the future have quest givers in towns. We also need locations to be more unique. Map size is good but some parts of the lands need further smoothing." But that's a lot of feature requests and suggestions to prioritize. Surely there's a better feature prioritization process. A quick COUNT(*) of desired feature options - using our categories feature - tells us that 40% of the people would like to see more pals added to the game! That's quite a significant portion of the reviews, and that’s just from people who bothered to leave one! Sounds like a good starting point 🙂 Since the data is locked and loaded on our end - let me know what else you think we should ask in the comments ⬇️ (and while you might not care about video games, imagine you could do this with your business questions) #videogames #gameproduction
-
Get textual data AI-Ready with Flexor Learn how you can incorporate UGC, like product reviews, directly into your analytics and AI workflows to drive business impact. Flexor takes care of the heavy lifting, so you can focus on moving the business needle.
Data that engages business leaders? Gamers (hey #GDC2024), retailers, e-commerce data folks: you're gonna like this. The Game Developers Conference starts today, so I decided to highlight one of the world's most data-driven industries 🎮 Even if video games aren't your thing, I bet you're all about engaging your BUs and driving business impact, right? Imagine you could base analyses on what users say, and back it all with numbers. --- Say Hi to Palworld 🧸 ICYMI - Within 8 weeks of its release, the game Palworld already has 282,593 reviews on Steam. It became the most-played third-party release on Xbox Game Pass, and it reached a peak of nearly 3 million daily active players on the Xbox platform. TL;DR - (a) Palworld players want more pals! (b) They like it (way) more than Pokemon. Curious how we learned this? see below ⬇️ To understand what people think of the game and how can this hype can be turned into dollars, we analyzed all of those reviews using Flexor. To put things in perspective, an average Steam review contains 57 words. That means there are 16,107,801 words, or over 32,000 pages. Since we at Flexor enjoy working with a lot of text, let’s dig in. First and foremost - it turns out the vast majority of reviews are positive. Remember the plagiarism accusations? Apparently, most reviewers think Palworld IS BETTER than Pokemon! However, we thought it would be interesting to ask - what can we do to make it even better? We loaded all the reviews to our Snowflake instance and ran the following SQL query: SELECT FLEX(review_id, “what features does the reviewer request?”) FROM steam_reviews_dataset Reviewers raised lots of good ideas, like: - "There is a ton of potential for expansions. A water / diving update, building update, new maps and Pals." - "There are some things I would really hope they add in the future: some semblance of a story, and actual shiny pals." - "Looking forward to seeing new Pals added in the future." - "In the future have quest givers in towns. We also need locations to be more unique. Map size is good but some parts of the lands need further smoothing." But that's a lot of feature requests and suggestions to prioritize. Surely there's a better feature prioritization process. A quick COUNT(*) of desired feature options - using our categories feature - tells us that 40% of the people would like to see more pals added to the game! That's quite a significant portion of the reviews, and that’s just from people who bothered to leave one! Sounds like a good starting point 🙂 Since the data is locked and loaded on our end - let me know what else you think we should ask in the comments ⬇️ (and while you might not care about video games, imagine you could do this with your business questions) #videogames #gameproduction
-
Flexor reposted this
Mid-2022, we were working on what would later become Flexor. Over 90% of the CDOs we interviewed said they don’t utilize unstructured data, because nobody in the business had requested it. Guess what happened last week? Another CDO has reached out: The CRO asked them why they don’t just “ChatGPT-their-way-through the CS emails” to alert him on churn signals. In retrospect, it wasn’t the CDO’s fault that he didn’t notice—it was a blindspot in 2022. Now? It’s discussed in every boardroom. To power your business with GenAI, you need the best data: clean, organized, well-structured, well-governed. And this is where Flexor comes in. If you are in charge of your organization’s Data & AI strategy and looking to incorporate GenAI across your data solutions, we should talk. Feel free to DM me directly. #datastrategy #aistrategy #dataleaders