minusxai (@minusxai) header
minusxai (@minusxai) profile picture

An open source agentic BI built around a SOTA data agent. Best agent + best looking plots! Backed by @ycombinator

San Francisco
Joined June 2024
Photo by minusxai (@minusxai)Photo by minusxai (@minusxai)Photo by minusxai (@minusxai)Photo by minusxai (@minusxai)Photo by minusxai (@minusxai)Photo by minusxai (@minusxai)Photo by minusxai (@minusxai)Photo by minusxai (@minusxai)
creating data stories was one of the great joys of my past life. fun to get back to it, esp with my man claude at my side! we will have more situations for you to monitor soon!
~25% of all seaborne oil flowed through the strait of hormuz until this war began. traffic right now is ~96% lower than usual. the deal maybe signed, but the real metric is if the ships start sailing. checkout this data story covering the ongoing chokehold on this critical world infra, as it unfolds.
1
1
3
441
~25% of all seaborne oil flowed through the strait of hormuz until this war began. traffic right now is ~96% lower than usual. the deal maybe signed, but the real metric is if the ships start sailing. checkout this data story covering the ongoing chokehold on this critical world infra, as it unfolds.
2
3
2
6
3,439
this is fantastic! if you want what ramp did inside your company, @minusxai can help! :) remember, llms cannot do magic. if you shape your data a bit and provide enough context, you can get a pretty magical data experience!
In the last four weeks, @tryramp Research (our AI analyst agent in slack) has answered 1,476 questions, compared to 66 in our help data channel. This is one of my favorite projects that we've shipped. Our team built an agentic solution based on tools to index domain docs, to search tables and columns, to run SQL iteratively: reasoning through data challenges the way a human would. As @bennstancil wrote: data's constant presence in an organization is like knowing the count of the deck. Though it makes us a bit more informed in each decision, the effect is only felt in the aggregate, as the small edge compounds over time. We are making a lot more data-driven decisions at Ramp than we were two months ago!
1
6
768
how is @claudeai's claude code so damn good and how can you recreate its magic in your own llm agent/workflow? intercepted all cc calls over the last few months and wrote up a 2k word guide on this! (all prompts, tools linked in comments) main takeaways (0/3): 0. debuggability >>> everything else. most of the magic is in designing good (low and high level) tools and prompts to let the model shine. keep it simple
12
33
12
178
113,915
our customer wanted to use @minusxai so bad we had to migrate all their queries and dashboards from redash to @metabase! So much so we made a repo for one click migration: github.com/minusxai/migrate-… hmu if you want to move!
1
6
639
i can remember things now! 😬
minusx memory: minusx md i love using claude code and one of the main reasons it is so good is the ability to just state your preferences via claude md file. but this is totally absent in analytics, where the problem is 100x worse - custom definitions, custom metrics, specific slices you want to see, etc.
1
497
gui gui gui @metabase
1
3
10
776
new release time! 🚀 Dashboards are where data-driven work begins. I spend a lot of time on dashboards and if I want to dig in, I’ve to copy SQL and head over to a query page and work through it. If I have custom metrics or dimensions, it is pretty much gg. We fixed this on @metabase! You can just ask @minusxai questions, and get answers, directly on any dashboard. Wrote about it in a blog post: minusx.ai/blog/dashboard-qa/ FYI, you can add this to your Metabase in <5 mins!
1
10
651
How many of the votes are *really* up for grabs in today's presidential election? One way of looking at it is seeing which counties remained with 1 party vs switched parties in the last 6 elections. Crazy that it is basically ~30-30-30 split.
1
3
611
Good morning messages I love waking up to. :) @apoorv_umang
1
10
758
Our goal with @minusxai is to give you an AI analyst that sits with you 24/7 in all your data exploration, analysis and presentation - wherever you are! Super super thrilled to add @googlesheets to our arsenal today! You can use MinusX on sheets! 😬
2
6
666
Imagine having an AI data analyst in your Sheets! Excited to announce that MinusX now adds on to Google Sheets with a simple one-click install: workspace.google.com/u/0/mar… If you use it & have any feedback/suggestions, please DM me! Here's what you can do right now:
3
11
522
You can use MinusX to fix any errors in the notebooks! Or ask for clarifications if something is still complex! 👻
This was us haha. @ppsreejith_ @arpit_tarang and I built @karpathy-fy! It takes a paperswithcode link, your current proficiency and gives you a tailored 5 lesson plan to go from where you are to the meaty parts of the paper implementation - like the karpathy video series / labml style. This is a task none of the previous models could do, and O1 was really good at. The best part? The output is an executable ipynb notebook (which you can use @minusxai to work through :) ) The below video has apple's recent monocular depth paper. I could actually go through and understand it, with loss functions from the paper and everything!
5
983
Error-shmerror. Fix it!
One of my favourite things about using @minusxai is that it makes all my Jupyter Notebooks self-healing! All of these can work on your own notebooks, like today! :) 1/4
2
973
We just open-sourced the entire MinusX chrome extension on GitHub under MIT License! The three main motivations for this decision were: Trust, Extensibility and Incentive Alignment. Read more about it in our blogpost: minusx.ai/blog/foss
2
2
2
13
1,676
Extensibility: We imagine almost all software we use today to either be replaced, or retrofit with AI. The retrofitting will not always be done by the makers of the original software. We're inspired by @comma_ai and even have adopted their opendbc folder structure! ❤️
1
1
5
1,030
Incentive Alignment: We don't want to lock you in, we don't want to sell your data. We want you to have the best data science assistant sitting with you. And when you win, we win.
6
375
Try this analysis out on your own models and see how they perform! Use MinusX to do it! :)
Replying to @nuwandavek
I did this whole analysis on Jupyter using @minusxai (in ~8.5 mins). If you use Jupyter (or Metabase) for analytics, give MinusX a try! Check out the blogpost for more analysis, the notebook and data: minusx.ai/blog/input-vs-outp… 5/5
1
481