Anania is a Robot data analyst. You can ask questions about your data in plain English, and get answers instantly! Anania has a Generative AI system which turns your questions into SQL queries and Python code, executes them on your data.
👋 Hey Product Hunters!
I’m Hrant, founder of Anania. We are super excited to go live on PH for the second time. We were surprised by the massive demand and waitlist sign-ups when we first launched Anania in February (1st product of the Day!).
Anania is a Robot data analyst. You can ask questions about your data in plain English, and get answers instantly! In behind, Anania’s Generative AI system turns your questions into SQL queries and Python code, executes them on your data, and provides you answers as tables, charts, or narrative text. We have launched a direct sign-up link this time to ensure everyone has instant access to our robot data analyst.
🔖How it started, and why?
We are Marvel fans. We love Avengers. We believe Jarvis did a great job rescuing Tony Stark from difficult situations. Our goal is to make Anania the assistant for real-life Tony Starks. We believe that any decision-makers should have the opportunity to get answers to their questions as fast and easily as possible.
We believe data analytics with plain English Q&A is the first cornerstone, but we can do much much more with Anania and with community support!
🚀 How it works
No setup, no configuration necessary. It is just 3 simple steps:
- Step 1 - Connect your data to Anania
- Step 2 - Start asking questions
- Step 3 - Obtain the results in numbers, texts, tables, or charts.
In behind, when you perform Step 1, Anania is fine-tuning its AI model to learn your data and be ready for your questions. After Step 2, it uses its knowledge about your data to translate your question into Python or SQL code, execute it and give you answers.
🎩 Awesome Features
We decided to make our features sound fancy and came up with a S.A.V.E. acronym 😃
- Search - Ask data questions in plain English to find insights from your data.
- Analyze - Calculate averages and ratios, find trends and correlations. Analytics made easy.
- Visualize - Get interactive charts and visualizations and export them when needed.
- Extract - Extract your results in your favorite format to create data-driven reports.
It is actually an early MVP, so we decided not to exhaust the product with many features yet, instead, work on improving our models. But we have ambitious ideas on what to add next!
🚥 There is more to come!
Here are some features that our research team is working on (we have prototypes!):
- Automated Daily Insights - we plan to call them “Anania Stories”
- Image analytics - ask questions on photos such as “how many images have red Panama hats?”
- Voice commands - speak to Anania instead of typing the question
- Store and Share - invite your team members to share your insights with them
- Adaptive AI - Anania can learn from its mistakes as well from the user’s input. we plan to call it “Teach Anania”
😍Thank you and thanks to our early adopters!
Our early adopters helped us tremendously to improve Anania! We still have tons of stuff to do. Join us, and use Anania to solve your problems. Find bugs and tell us so that we can fix them for you (and for us)! Specifically for ProductHunters, we have a specialist coupon code that you can use to get our Basic plan with 50% LIFETIME discount! 🎁 Just add “AnaniaPH” as the Promotion Code when checking out and enjoy it!
@dan_robins hey Dan, we are currently providing file upload and relational DB connection. Next up are CRMs (SalesForce and HubSpot), Data Warehouses (Snowflake and RedShift) and then we are thinking about either Shopify or AirTable.
@hrantdavtyan Sounds like a great roadmap! I was wondering whether you might also look to integrate with some tools like Miro and Mural (seems to be where a lot of teams are storing data and collaborating - but notoriously difficult to search through!)
@artem_harutyunyan Hi, thanks for the question. Currently we support CSV/Excel file upload and PostgreSQL DB connection. Other sources are available per request.
Hey, congrats for the launch! I was wondering, is the generated SQL efficient enough to work (in reasonable time) on very big databases ? My fear is, if you are using generative model, you might have a problem where the SQL generated works in theory but won't in practice due to size of the DB. Or is it already efficient enough? Do you use the size information in the prompt to the generative model? What are your results on this matter?
Thanks!
@gregoire_ret that's a key question thanks, especially from our Enterprise users!
at this point, we optimized the model to generate as optimal SQL as possible, but we will never have full control. we are planning to integrate a 3rd party tool, which measures the efficiency of SQL before execution and automatically makes improvements. that is on the Q1, 2023 roadmap
@hrantdavtyan Amazing! Yeah I feel like a post-processing approach by another tool is the most reasonable solution as "you can never be sure with DL". Thanks for answering :)
Replies
Anania
Anania
Anania
Anania
Anania
Anania
Anania
Anania
Anania
Anania
Anania
Anania
Anania
Anania
Anania
Anania
Anania
Bardeen
Anania
Anania
Anania
WayAway
Anania
Free PR service powered by ChatGPT
Anania
Anania
Anania