Pipo AI
Introducing Pipo AI: A new tool for automating data pipeline generation (ETL) and maintenance by analyzing input-output examples to handle transformations automatically.
For example, a Data Engineer can use us to connect easily the input of Stripe to Snowflake without writing any code.
🎯 Motivations
Doing transformation is one of the most time-consuming task for engineers: it is different for each integration you want to implement.
Complex transformations can be a mess.
The maintenance of the data piple can be hard and involves a lot of people.
We won't lie: it is a boring task!
✨ Features
- MistralAI powered:Use the power of Mistral 🇫🇷 to generate an optimized code
- Open-Source: Let the world use Pipo AI
- Schema validation auto-generated: Infered json structures from your inputs/outputs
- Auto mapping: Reformat your data to match with the desired output. It can handle simple transformation (like create a
full_namefield fromfirst_nameandlast_namebut also complex transformation like date/time format transformation) - Auto maintenance: Automatically detect and add new columns created
🗺️ Roadmap
This is an early project but we have already these features in mind:
- Support different input/output formats
- Self healing pipeline which ping you on Slack when your schema is not anymore updated
- Act as Segment: be a proxy to different sources (authentication and security)
- Connection to different input and output sources (DBs, API)
- Use openapi spec as validation schema
Built With
- mistral
- python
Log in or sign up for Devpost to join the conversation.