Technology Panel: Big Data and Machine Learning

In its biggest and most diverse event to date, Women in Big Data’s™ North West Chapter Networking Committee presented Technology Panel: Big Data and Machine Learning to an enthusiastic audience of 150 people. The event took place August 9, 2018 at Intel’s Hawthorne Farm Auditorium in Hillsboro, Oregon, and the topics covered included technical challenges, bleeding edge technology and steps to take to enter the field. The event was also well attended by men (~at least 25%).

A great audience

Technology Panel was a great platform to network, exchange ideas and explore a hot topic from diverse perspectives. Panel speakers represented an outstanding combination of business acumen and technological experience. Speakers were diverse in terms of thoughts, experience and ideas. In the course of the evening, we saw senior technical leaders emphasizing the importance of technical expertise and how each of us can shape our career in a world where change is the only constant. We also had panelists from the startup world discussing key skills needed to survive as an entrepreneur, and others who leveraged their background in data management of critical cloud products.

This broad line of speakers fueled the minds of our audience with technical insights and advice. It also provided a great medium to ask questions, with the panel fielding nearly two dozen substantive questions before time ran out.

L. to R.: Bhakti Hinduja, Agata Gruza, Kelly Hammond, Sherine Abdelhak, Richa Khandelwal, Parham Parvizi

The panel discussion kicked off with an a welcome by Soumya Guptha and an introduction to Women in Big Data (WiBD) by Bhakti Hinduja, who shared a profound quote form Sheryl Sandburg: “In the future, there will be no female leaders. There will just be leaders.”

The actual panel was divided into two segments: a brief introduction of all panelists with Q&A from moderators and then Q&A from the audience. The goal of the panel was to provide the audience with broad horizontal knowledge of Big Data/ML from many different angles. Questions were tailored to each panelist’s expertise.

 

Richa Khandelwal, Sr. Software Engineering Manager @Nike received questions related to the software industry. Some of them: What is the greatest challenge you have faced when rolling out first Machine Learning applications to consumers? DevOps aiming to unify software development (Dev) and software Operations (Ops)? What are best known practices to shorter software development cycle? How to be relevant? How to make others listen to you?

 Parham Parvizi, Co-Founder and Architect @Tura.io received many questions related to opening your own business. Some of those questions and topics: How to evaluate if a startup idea is worth the effort? Top skills/qualities that are needed to thrive in the startup ecosystem? How to determine if the startup ecosystem is a good choice for oneself? Is timing a key factor?

Kelly Hammond, Open Source Technologies Manager @Intel received questions related to effectively managing artificial intelligence (AI) projects. Some of those questions: What is the next bleeding edge technology in Big Data/ML? Where do you think Big Data will be in 10 years? How do you measure the success of project and engineer in Big Data/ML? How much about Big Data/ML do I need to know before making a permanent transition into Big Data field?

Sherine Abdelhak, AI, Platform Architect @Intel received questions related to AI for client platform and transitioning between two different fields. Some of the questions: Are we preparing our platforms for AI, and how? Why is AI applicable to client? Are there specific AI technologies that client predominantly might benefit from?

Key takeaways:

  • Big data is more than just lots of data. We have to think of analytics to be predictive. AI is your friend—it can help you find patterns in the data. Key is moving from understanding what happened to understanding what will happen and what we can do about it.
  • Not size but quality of data is a key. But how do you validate that data are not corrupted? You need to treat it like gold.
  • What does perfect data look like? Right data are consistent. We know there are right when we are collecting the key metrics. The timeline of collecting the data is crucial.
  • Big Data is expensive. Reducing the cost of Big Data/AI is key.
  • What are some good and big enough use cases for a Big Data project? Providing better recommendations is a big use case. How do we mitigate bugs and issues in Big Data? How do we identify and solve problems faster is another use case. Genomic sequencing. Supercomputers vs AI. BigDL as an Intel Open Source AI framework. Machine Learning and Reinforcement Learning come in handy and can help tackle those problems.
  • What are Big Data use cases for social causes and non-profit organizations? One example is demographic analysis at scale.
  • What does a data scientist do all day? Lots of learning about new technologies and knowing what competition is doing. In addition, learning new ways how to communicate those findings to others. There is lots of overlap between data scientists and engineering managers. Lots of data scientists and engineers are coming to AI.
  • AI is converging into the Hadoop/Spark ecosystem. There are many applied mathematicians involved in AI, but everyone can learn AI.
  • Reinforcement learning is a type of Machine Learning (ML) that act on the prediction. Moving from insights to prediction is one of the next big things. Invest your time in it.
  • Investors first look for what is the market, is it big enough and what problem are you trying to solve; do you have the right solution to that problem; are you the right person to do the job; are you committed; do you really have the passion and know-how; are you broadening your understanding of what AI and Big Data are?
  • What is one piece of advice that you would like the audience to remember from this panel? Failure is a good thing because we learn from it. But learn fast. Don’t focus on what you know but on what you can learn. Just do it—find a problem you can solve with big data, AI and ML. That way you will be in the forefront of the field. Build your hypothesis and challenge it fast. Work efficient. You have the confidence and passion to learn, so learn AI and data science because it’s the future.

The Technology Panel wouldn’t have happened without a fantastic team working behind the scenes. Many thanks and congratulation to:

L. to R.: Bhakti Hinduja, Mangai Vetrivelen, Soumya Guptha, Agata Gruza, Khanak Nangia, Sherine Abdelhak, Kelly Hammond, Richa Khandelwal, Parham Parvizi, Sravani Gomatam

Many thanks to all panelists, who in a coherent way educated, energized and catalyzed enthusiasm for getting into Big Data and ML. Thank you audience for warm and encouraging feedback. Some of their comments:

“This was such a great event! Heard so many perspectives and learned about what inspires people. Thank you very much for organizing!”
Kiril Simov

“Yes, I would recommend because it helped me stay updated and learn more about AI and ML. Despite the fact that Intel is shifting its focus to become a data-centric company, I believe AI can also be used to improve our internal operations as well.”
Anonymous

“Great event! I hope to see and hear more about big data in the future.”
Joanna Tan

The number of attendees and the positive input from participants confirms we are doing important work and there is a need for similar events.

 

Leave a Reply

d
c