Join Sammy Warrayat and Sol Girouard as they discuss blockchain and data modeling. Sol is a mathematical economist, data scientist, quant, and the CEO and founder of Data Innovations Lab. In this episode, she sheds light on what blockchain really is and everything they do related to machine learning.

What Really Is Blockchain?

Almost everyone equates blockchain to cryptocurrency, but Sol says otherwise. Blockchain is the base for everything known as distributed ledger technology (DLT) or decentralized ledger. Cryptocurrency is just one of the functions that can be given to a blockchain. Sol says that in a company, it is very smart to have a blockchain to work on a clear function or to add multiple functions per ledger because, in that way, it is clean. A lot of good usable data can also be taken from it without doing extra data processing. There are private and public chains, and whatever happens on one side must also happen to the other side, just like an accounting ledger. It also has a consensus that nodes are peers or blocks, so they have to say and approve the identity of whoever does the transaction, which makes excellent use of blockchain for identity protection. The level of security differs between private and public chains, but both have benefits and uses depending on the functions. 

Domain Knowledge

Blockchain is only one component, but not all data go into the chain; there have to be other data sources. A data scientist cannot work on every data set because they are not knowledgeable about it, so Sol highlights that domain knowledge is necessary. Data science is a field of intersection and being proficient in coding algorithms, deep learning, and knowing statistics is not enough. A data scientist should also have expertise aside from data science itself as it solves clear points. All of this comes to play. If the data scientist does not know where the questions are assessed from, they cannot perform the models properly nor acquire the skills needed to make them intersect. Having no knowledge from a field the data is extracted from results in wrong insights. Thus, an area of concentration is required even if one can handle large data very well.

About Sol Girouard:

Sol graduated top of her class from Harvard University and holds Academic Teaching Fellow positions for Data Science courses at Harvard’s Engineering School—SEAS, Institute of Applied Computer Science, and Harvard Extension School. She has judged the University of Chicago Financial Mathematics Master Program Graduating Competition.

Sol is the CEO and founder of Data Innovation Labs, a full-service data science and decision intelligence consulting group deploying 4IR technologies and implementing digital transformation across business verticals. She is also Blockchain APAC venture Laclary’s founding partner and holds Data Science Advisory positions in international financial and 4IR firms. Sol was the chief economist and head of Fundamental and Quantitative Research at Trevinci Capital Partners—liquid Ag Hedge Fund. She was the managing partner and head quant for Oracle Management—a proprietary global macro trading firm.

Outline of the Episode:

[0:02:45] Sol’s introduction about herself and her journey

[0:06:51] What was it like going to Harvard and the issue of underrepresentation

[0:13:33] An ecosystem of companies built around machine learning and blockchain

[0:16:43] Distributed ledgers, decentralized banking, blockchain, and cryptocurrency

[0:22:40] Latency benefit that private chains have over public chains

[0:23:53] How blockchain nodes make life easy for data scientists

[0:28:35] The program that allows for continuous machine learning

[0:37:09] Sponsor advertisements

[0:39:34] Why should you care about artificial intelligence?

[0:42:26] The silver lining in Sol’s hearing disability&