EAIML #46: Answering Questions About Your Data, Thomas S. & Vladimir B., deepset

Important:
if you click on the Zoom link before 18:00, you will see on your screen that you are placed in a waiting room. This will disappear once the broadcast begins at 18:00 sharp.
Whenever you are ready, please feel free to enable your camera, to make it easier for you to connect with other participants!
Zoom link:
https://us02web.zoom.us/j/82427335473?pwd=am8yTk9VdTJHVjdqd29PaHJ5TUNNdz09
*
Dear AI & ML Enthusiasts,

This time we have an exciting talk on
Answering Questions About Your Data
by
Thomas Stadelmann & Vladimir Blagojevic, deepset GmbH.

Agenda:

Welcome

Interview

Talk:
Answering Questions About Your Data
Thomas Stadelmann & Vladimir Blagojevic, deepset GmbH
https://www.linkedin.com/in/thomas-stadelmann-47b71218b/,
https://www.linkedin.com/in/blagojevicvladimir/
Abstract:
With today's open-source frameworks, LLMs make question-answering systems as powerful as never before. However, with an overwhelming number of potential configurations, users often struggle to find the best configuration and prompts for their individual data.
This talk shows you how to find the best fit for your data by systematically iterating on prompts using a retrieval augmented setup. We take a look at prompt engineering, the role of retrievers and one of the biggest problems in today's LLM-based solutions: hallucinations.
We finish the talk with a glimpse at an advanced concept, that we might see more often in the future: agents.

➡️About the Speakers:
Thomas Stadelmann is Senior Software Engineer at deepset, where he works on Haystack, deepset's Open Source Framework for Semantic Search, and on deepset Cloud, their complementary commercial product. He got his Masters Degree with distinction in Information Systems from Otto-Friedrich University of Bamberg. He early focused on Information Retrieval, Search Evaluation, Query-Log Analysis and A/B-Testing. After working on domain-specific real-world search applications for over 11 years, he was thrilled when modern AI & NLP techniques reached the information retrieval realm, which led him to join deepset in 2021.

Vladimir Blagojevic is a Senior Software Engineer at deepset, working on Haystack, deepset's Open Source Framework for Semantic Search. Earlier in his career, he worked on the distributed system at JBoss/RedHat. After thirteen years at RedHat, he recently refocused on deep learning and natural language processing. He got his Masters's Degree in Computer Science at York University, Canada and joined deepset in June 2022.

Looking forward to seeing you all!

Alex

P.S.
➡️ If you are on Telegram, the best place to receive updates about the meetup is the Telegram channel
https://t.me/EAIML
You can also join the channel by searching for EAIML on Telegram.

➡️ If you do not have Telegram yet, you can get a Telegram Account for free here https://telegram.org

➡️ If you have further questions about the meetup, the best way to contact me is on Telegram. Click on https://t.me/alexanderlorz or search for alexanderlorz.

*************** IMPORTANT ***************
Photography / filming / audio recording is taking place during the event for documentation purposes. The photographs and recordings will likely be published for the purpose of reporting on the event on all communication channels of the event organizers. If you do not wish this, please inform the organizer on site before the start of the event.