Employee Spotlight: Zav Shotan

Name: Zav Shotan

Department(s): Engineering, Data Science

Joined: 2018

Hometown: Yehud, Israel

Spoken Languages: Hebrew, English

 

 

 

 


Joining the Cherre Family

When I finished my second post doc, I was mostly looking for interesting projects. Cherre’s ideas about how to approach data, both in the algorithmic sense of entity resolution and in the core ETL process, were refreshingly clear and direct. To me, the challenge of ML identification of prominent factors (companies, players, locations and even contracts) in an environment as “dirty” as real estate, seemed very appealing. On a more personal note, unlike other places, when I walked into the office for the first time most of the people in Cherre had smiles on their faces. The discussions we had were on point, concise, and exciting.

 

I’m a Snow Leopard

My work focuses on algorithms, ETL, data pipelines, and domain. I can combine the more specific and scientific algorithms of Data science into production and the ETL process. My background in physics where I dealt with multiple disciplines and systems allows me to identify some of the pitfalls when it comes to multidisciplinary integration.

I’m currently working on three interesting projects:

(a) ETL – create a template engine tool to allow the easy processing and integration of multi-language (SQL, Python, Javascript, etc.) code into our ETL scheduler (currently Airflow), all while allowing us to change between cloud providers (Amazon AWS, Google Kubernetes, etc.).

(b) Source Identification – the automatic processing of SQL (and other) code so that we can present to our clients what processes their data is traveling through, and identify pitfalls within the process. The aim is to produce graph diagrams, which later on will become interactive.

(c) Entity Resolution – given various data sources, we identify the cardinal objects, and assign each a persistent object ID where we can track and follow the entity progression with time. For example, with a building we can track all owners (each an identity in itself), all the work that was done, when it was built (born), and when it was demolished (died). We recently wrote a patent on this topic.

I’m also working on a side project related to 3D printed drones with machine vision: I am trying to make an amateur, 3D printed, drone that uses a Raspberry Pi head to perform object recognition in free space (based on common algorithms and leveraging TensorFlow). One day maybe it can fly on its own.

 

Changing our World

Since I am a pessimist, I do not believe we can change anything in time to save the current planet. I think space exploration and colonization is very important. On more general topics, the advancement of core science (Physics, Chemistry, etc.) has been almost abandoned to pressing issues, which in my opinion is an oversight.

 

Growing Personally and Professionally 

I’m currently reading Knowledge and the Wealth of Nations: A Story of Economic Discovery by David Warsh (Amazon).

 

Saturday Morning Fun

Bill Burr routines. 


The Narrator of Your Story

Bill Burr.