Fabi.ai

Fabi.ai

Software Development

San Francisco, CA 1,851 followers

The AI data analyst you've been waiting for.

About us

Fabi.ai combines SQL, Python and AI automation into one collaborative platform to help you conquer complex and ad hoc analyses, turning questions into answers.

Industry
Software Development
Company size
2-10 employees
Headquarters
San Francisco, CA
Type
Privately Held
Founded
2023

Locations

Employees at Fabi.ai

Updates

  • In 2024, we were hard at work making data analysis fun, easy and accessible to all. We wanted to take a moment to share with you some of the cool things that we’ve put out in the world, and give you a bit of a sneak peek into the future of data analysis. We’re unlocking new levels of efficiency with: 🤖 Our new AI agent which can now reason through multiple steps and automatically take action to install packages, dry-run code or retrieve task-specific context. 🐍 A mixed SQL and Python environment so that you can do 90% of your analysis in SQL, the language we know and love, and tap into Python for the extra gnarly wrangling. 🦆 DuckDB support to provide a snappy, flexible data analysis experience. And we’re making it easier than ever to share insights the way your team wants them with: 🔌 A DataFrame to Google Sheets connectors so you can just push data straight to a spreadsheet, and skip the “Export” button 🔔 Quick and easy ways to push alerts to Slack so that your stakeholders get timely alerts right where they work. 😎 Interactive reports that can get published in 2 clicks and scale with your data. No more deployment hassle for your data apps and analyses. 2025 is going to be the year where AI not only supercharges data teams, but changes the way they operate. Data teams are going to shift left with a renewed focus on solid data foundations, which will then allow them to put AI in the hands of the business with more confidence. This in turn will put data at the center of more strategic decisions, fueling growth for organizations that embrace this shift. And Fabi.ai will be the platform that makes this a reality. And if you love beautiful, intuitive interfaces and want to join us, we’re hiring a senior front end engineer! Happy HolidAIs from the Fabi.ai team! 🎄

  • Check out our walk-through of our newest AI agent! Data analysis just got a whole lot easier.

    View profile for Marc Dupuis, graphic

    Co-founder & CEO @ Fabi.ai | Building the AI data analysts you've been waiting for.

    AI agents are going to change the way data analysis is done. I know this sounds like hyperbole and AI agents just feel like the latest fad, but I assure you, they’re here to stay. Last week we announced our AI agent as generally available, and this week I wanted to share a video showing it in action a bit more (full credit to Lei Tang). Before having an AI agent: ❌ The AI had to get the code right in one shot. This is really hard to do (even for a human!). Say you ran some code and got an error, but there’s actually another issue lurking a bit further down in the code... The AI may be able to correct the first issue, but you’d quickly run into the second issue, forcing you to debug multiple steps. ❌ The entire context had to be jammed into a single prompt. We used RAG, but even then, you had to get _just the right_ context for the AI to produce good results. Too little context and the AI will provide incorrect info, too much info, the AI will get confused and hallucinate. ❌ The prompts being so large, the AI response time would inevitably be slower, making for a poor user experience. With the Fabi.ai AI agent: ✅ The AI is given a set of tools that it can invoke to complete tasks and those tools retrieve just the right context for that specific task. For example if the AI first identifies that the question can’t be answered with the data in the existing analysis, it can invoke a function to retrieve specific DB schema information. ✅ Another tool that the AI has is “dry running”. So when it produces code, it can check that the code runs, and if it runs into an error it will continue debugging. Thanks to this, it will get much closer to the end results on its own than when it uses a one-shot approach. ✅ Individual prompts for each function are much more specific and much smaller. This both increases accuracy AND decreases latency. This is a huge win for everyone. Things that make our AI agent even more special: 🔀 We’ve built this to be completely AI agnostic. You can use any of the major providers, or even a Fabi.ai-hosted LLM. 💻 There’s some very clever and sophisticated kernel and virtual machine management happening to make the whole experience “just work”. Check out the video, or take it for a spin yourself 👇

  • Our mission is to make data analysis and distribution of insights across the team incredibly easy. In November, we made massive updates to help you do just that. Now with our AI agent designed for data analysis, AI can reason through multiple steps, helping you uncover insights and build customized visualizations even faster. And with our flexible reports layouts you can build interactive dashboards in seconds to share those insights. 🧑🎨 Customize your report layouts Turn your awesome insights into fully customizable dashboards and reports in seconds. Add or remove tables and charts, move things around, customize your layout to your hearts content. 🤖 AI Agent (Generally Available) Our AI agent is out of beta and generally available. How does it work? Give the AI an objective and some tools (functions) to complete that task and let it work through the task. The first iteration of our AI agents auto-tests the code and can fix multiple code issues at once, helping you get to your answers faster. This is perfect for more reliable AI outputs and AI auto-debug. No more “That resolved my first error, now I’m getting a new error…” The future of AI for data analysis is agentic and there’s a lot more to come from us on this front. ⏰ 100% customizable report schedules Want to schedule your report or Slack alerts to run every other weekday at 5AM and 5PM unless it’s a full moon? No problem. Well almost… cron jobs don’t yet track lunar cycles. Just ask the AI and it will write the cron expression for you so you don’t have to. Our complete November product release notes available in comments 👇

  • 🔥 Hot data jobs! Featured job of the week: Senior data analyst Typeform Why we think this job is cool: 1️⃣ Typeform is a cool product changing the way forms are shared and data is collected 2️⃣ You would be working closely with the GTM team where your impact would be greatest 3️⃣ They’re working with a modern data stack Some of the other jobs feature by: Mike Angelo hiring at Typeform Lyndsey Lustig hiring at Headspace Franciska Dethlefsen hiring at Amplitude Ben Straley hiring at Kigo Liya Aizenberg hiring a Major League Soccer Kory Doran hiring at AppFolio Sam Myers hiring at Leaf Home Tim McWilliams hiring at Ovative Group Full list in our data job directory linked in comments 👇

  • In this tip of the week we show you how to turn a static report into a dynamic analysis for your stakeholders 🎚️ If you regularly do ad hoc or exploratory data analysis, you understand the need to turn your work into something that your stakeholders can “play around with”. We make that incredibly simple with filters and inputs. In just a few seconds, you can turn any SQL or Python analysis snippet into a dynamic report with: 📅 Date and date range dropdowns 🔘 On/Off or True/False boolean toggles 1️⃣ Number and number range selectors 💬 Arbitrary text inputs ☑️ Single or multi-select dropdown lists Check out our short video to see it in action 👇

  • Data job of the week: Vouch Insurance 🛡️ - Senior Data Analyst Why we think this job is cool: 1️⃣ Vouch is a strong and growing insurance brand with a focus on the tech industry 2️⃣ They have offices in major metro areas across the US 3️⃣ Lots of great perks and clear hiring process outline and expectations Key skills you should have to apply: 1️⃣ Advanced SQL skills, experience with BI tools (Mode, Tableau, Looker), and familiarity with data transformation tools (dbt) 2️⃣ 3-6 years of experience working as a data analyst, data scientist, consultant, or equivalent quantitative analytical role This week's new jobs include jobs from Aurora, PandaDoc, The Guitar Center Company and Zapier. Full job board in comments 👇

  • View organization page for Fabi.ai, graphic

    1,851 followers

    Your first data hire can change the trajectory of your business 🚀, or… make you completely question the value of data 🫠 When are you ready for your first full time data hire? 📊 You already have in place basic reporting and you have some data (no matter how clean or messy) to report on. 🧭 You’re ready to use data to adjust your plans and inform you decisions. 💵 You’ve budget for the tooling to support this data person, in addition to their salary. What should that person’s profile look like? 📈 Ideally they should have experience building and scaling data teams. At the very least that should have been mentored by someone who has. 🐍 They should have extensive SQL and Python experience. 🛠️ They should not be be singularly obsessed with a specific toolset that they can’t live without - Initially they will need to be scrappy, you can’t wait 3-6 months to have your first reports stood up and they may need to work with what you already have in place. 🧮 They should have a strong desire and ability to understand the P&L of your business and communicate effectively at the leadership and board level. When you hire this first person, it’s easy to think that you need a data scientist to uncover magical insights or perhaps you can hire someone more junior and affordable who knows their way around SQL, but  these decisions often come at the success of both parties. In our latest post, we sat down with Aditya Goyal, Senior Director of Data at Shogun to talk about what it takes to build the right foundation for a successful data team. Full post in comments.

  • Fabi.ai reposted this

    View profile for David Hyde, graphic

    Analytics | Sunobi

    After Saturday's emotional Utah-BYU rivalry game (yes, there were tears), I leveraged Fabi.ai's notebook platform to analyze NFL replay trends. Seeking comfort in data (as we analysts do), I dove into 4 years of NFL history. Key discoveries from the data: - Replay reviews impact 7% of plays (they are a big deal) - 57% reversal rate across all reviews - Location matters: Red zone sees increased reversals - Timing is everything: Q1 reviews most likely overturned - The eternal question "Was it a catch?" helps drive 73% of reviews coming from passing plays - Team reversal rates varies widely: Cardinals/Rams (73%) vs Falcons (44%) - Counterintuitive finding: Reversed calls ≠ better outcomes Full analysis available here: https://lnkd.in/g3Q74EWc #SportsTech #Analytics #DataScience #NFL

    • No alternative text description for this image
    • No alternative text description for this image
    • No alternative text description for this image

Similar pages

Browse jobs

Funding

Fabi.ai 1 total round

Last Round

Seed

US$ 3.0M

See more info on crunchbase