Harnessing Data Magic

Let a Genie Find Your Needle in a Field of Haystacks!

QUEENS NY DATABRICKS USER GROUP

MEET YOUR HOST

Daniel "Danny" Lee

Technology Lead, Infosys Ltd.

Data and Analytics (DNA) Specialist�Domain: BFSI Regulatory Compliance

Learn & Share

Solve Challenges

Network

QNYDUG Group Aims

To learn and Share more about:

Data Engineering

Databricks

Queens, NY

Help solve real-world challenges

Introduce people and network

Links

Official Databricks Links:

Databricks Data+AI Summit (Virtual): https://www.databricks.com/dataaisummit

NYC Databrick Community Meetup NYC: https://www.meetup.com/databricks-community-meetup-nyc/

Databricks Free Edition: https://www.databricks.com/learn/free-edition

Databricks Community Forums: https://community.databricks.com/

Once you’re there, say hello and connect with me — my username is Danny_Lee.

Our DUG: https://usergroups.databricks.com/queens-new-york-databricks-user-group/

Databricks User groups – create your own: https://usergroups.databricks.com/chapters/

Databricks Release Notes: https://docs.databricks.com/aws/en/release-notes/product/

Databricks YouTube: https://www.youtube.com/@Databricks

​

Group Links (unofficial)

Our DUG LinkedIn: https://www.linkedin.com/groups/17534047/

Our DUG Meetup: https://www.meetup.com/queens-ny-databricks-user-group/

Our DUG Luma: https://luma.com/qnydug/

Our DUG YouTube: https://www.youtube.com/@qnydug

Our O'Reilly Book Recommendations: https://tinyurl.com/3h2s4d3z

Speaker Proposal Form: https://forms.office.com/r/GABbiPh5a9

QNYDUG Zoom background: https://tinyurl.com/qnydug-bg

​

Dataset Sources

Kaggle - https://www.kaggle.com/datasets/snap/amazon-fine-food-reviews/data

NYC Open Data - https://opendata.cityofnewyork.us/

US Gov Data - https://data.gov/

​

Appendix:

Recommendation: Don’t choose passkey. My experience has been that its not convenient for logging on from different places (phone, laptop, ipad, etc).

THE CHALLENGE: ANSWERS ARE SLOW

"We have the data... but current tools aren't agile enough."

Huge Datasets

Enterprises are drowning in data. Finding specific answers across billions of rows manually is like searching for a needle in a haystack.

Analyst Bottlenecks

Every simple question turns into a SQL request. Business users wait days for answers while experts are buried under routine queries.

Dashboards Limited

Fixed visualizations can't answer "why" or "what else". Users are stuck with pre-built views that lack real-time conversational depth.

INTRODUCING DATABRICKS GENIE

A Generative AI-powered conversational analytics interface that democratizes data access across your organization.

Natural Language: Ask questions in plain English.

Autonomous SQL: Auto-generates complex queries.

Instant Output: Tables and visualizations on demand.

THE MAGIC WORKFLOW

1. Input

User submits a natural language question about the business.

2. Logic

Genie translates intent into a precise SQL query via Unity Catalog.

3. Action

Query executes against your Lakehouse to fetch live data.

4. Insight

Results are summarized and visualized for the user.

THE DATASET: FINE FOOD REVIEWS

Amazon Fine Food Reviews

A massive corpus of user sentiment containing ~568,454 reviews across several years.

Genie Demo

Databricks Free

Create Volume

Download data from Kaggle

Load data to Delta Catalog

Interact With Genie

Curate and Guardrails in Genie

Genie Demo

SCENARIO 1 Top products by rating

SCENARIO 2 Trend over time

SCENARIO 3 Find rating drops

SCENARIO 4 Low rating but helpful

JOIN THE QNYDUG COMMUNITY

Official Resources

• Data+AI Summit: databricks.com/dataaisummit�• Free Edition: databricks.com/learn/free-edition�• YouTube: youtube.com/@Databricks

Our User Group

• Meetup: meetup.com/queens-ny-databricks-user-group�• LinkedIn: linkedin.com/groups/17534047�• Luma: luma.com/qnydug

Questions?

Connect with me: Danny_Lee on Community Forums

https://community.databricks.com

See you next time!