Elastic OTEL & Cloud Serverless + ETL for RAG with Aryn DocParse & Sycamore

The Elastic Developer Relations Team will be in town sponsoring PyTexas, and we will be hosting a meetup before the conference on Thursday, April 10th, 2025.

We'll have presentations from Jessica Garson (Sr. Developer Advocate at Elastic) and Dan Tecuci (Head of AI at Aryn), followed by networking, pizza, and refreshments.

📅Date and Time:
Thursday, April 10th, from 5:30-7:30 pm CDT

📍Location:
Elastic Austin Office: 8th Floor
823 Congress Ave
Austin, TX 78701

🚗 Parking:
Parking options nearby include several garages or street parking in nearby blocks. Check out SpotHero to reserve a parking spot in advance!

🪧 Arrival Instructions:
When you arrive at the office, head to the 8th floor.

The main entrance to the building is on 9th Street, between Congress Ave and Brazos St.

📝 Agenda:

• 5:30 pm: Doors open
• 6:00 pm: Document ETL for RAG and Semantic Search with Aryn DocParse and Sycamore - Dan Tecuci (Head of AI at Aryn)
• 6:30 pm: Integrating OpenTelemetry with Elastic Cloud Serverless - Jessica Garson (Sr. Developer Advocate at Elastic)
• 7:00-7:30 pm: Networking

💭 Talk Abstracts:
Document ETL for RAG and Semantic Search with Aryn DocParse and Sycamore - Dan Tecuci (Head of AI at Aryn)
It’s critical to properly prepare unstructured data when building RAG or semantic search applications with Elasticsearch. Creating the proper ETL pipelines with document segmentation, table and image extraction, OCR, data enrichment, data cleaning, and more is not trivial when dealing with complex data. In this session, we’ll show how to build advanced document ETL pipelines with the open source, scalable Sycamore library and use Aryn DocParse for critical processing steps.

Integrating OpenTelemetry with Elastic Cloud Serverless - Jessica Garson (Sr. Developer Advocate at Elastic)
This talk will walk you through how to quickly set up an application to use OpenTelemetry. OpenTelemetry is an open-source, vendor-neutral way to add monitoring features to your application. It is designed to allow you to monitor different systems using different backends in a standardized way. Using serverless, you can get started with Elastic quickly without worrying about your cluster's usual maintenance.