Agenda

Main Stage: Chair's Opening Remarks

Alex Weber Award-winning host, motivational comedian speaker .

Main Stage Opening Keynote: Virtual Beings Not Virtual Assistants

Edward Saatchi CEO Fable Studio

Main Stage: Modern Day Technology

- What AI means for potential business growth, examples & use cases

- Lessons learned through the definition of the Artificial Intelligence strategy

Dan Devone TV Personality NBC Sports

Generating real value from real time analytics

The value of data and its analysis is often highest in the moments after it’s created. Real-time analytic solutions unlock this value by driving data into real-time decision-making and business processes. Join this session to learn how Google Cloud and its Data Analytics Platform can help simplify your analytics workflows and allow your teams to focus on getting to insights from your data for your industry and your customers.

Evren Eryurek PhD Director, Product Management Google

Evolution of Data Science at Reddit: from ML Insights to Experiment-Backed Data Products

Success of data science requires mastery of statistical/ML techniques, and deep understanding of company priorities. The most rewarding yet challenging aspect is to build a rhythm to deliver actionable ouputs (i.e., data products) that directly transform business decisions. Here in Reddit, we have uniquely complex data assets, for which it took us rounds of iterations to customize a solution that maximizes our data potential. This talk focuses on our recent journey in the last 18 months to build an experimentation platform, and to revamp the engineering workflow prioritizing the interplay of offline ML modeling and online testing. Not every piece of ML insights checks out in an experiment, but every "setback" is equally valuable as it allows us to iterate faster and more purposefully. Culturally speaking, we encourage calculated risk taking since data products are most powerful when they point to new business directions (vs. validating the existing ones). With a suite of concrete examples, we demonstrate how rhythmically outputting data products answers the common questions that almost all companies are curious about, namely, where are “good ideas” from? How to generate them at scale? Do they hold up IRL? Still valuable if they don’t?

Lin Huang Director of Data Science & Engineering Reddit

How simulation is transforming advanced analytics

Your data management initiatives are finished. The analytics team is in place. Business users are satisfied with their visualization and BI tools. But what’s next? The answer: Simulation. This session shows how simulation addresses the complex dynamics of any market, including consumer behavior and competitors, to deliver trustworthy answers to business questions. Learn how leading organizations are integrating simulation into their advanced analytics stack to answer their what-if business questions.

John Pasinski VP of Analytics Concentric

Lunch

With innovative launchpad presentations from:
Nico Rode, TIBCO
Wade Tibke, Sigma Computing
Michael McNair, Codey.ai & 55B Labs

Enterprise-Scale Innovation that Delivers Business Results

The presentation will focus on best practices to develop ML powered applications that can move the needle on business critical KPIs. We will walk through a rapid prototyping framework to develop effective personalization experiences, the mindsets and skills required to execute on an innovation roadmap, how to evaluate and work with vendors that provide 'AI-powered' solutions, and how to design experiments to quickly iterate towards a better experience for customers.

Pallav Agrawal Director, Data Science Levi Strauss & Co

Building a Data Driven Organization in 3 Easy Steps and 1,645,242 Less Easy Ones

Data is all around us and is a driving goal for many organizations, but while the goal may be in sight, the steps needed to achieve data driven results are not early as clear. Trying to do too much at once or being pulled in to many directions is a potential roadblock to getting good repeatable results that allow you to gain solid business insights. This presentation hopes to provide a roadmap that outlines the high level steps needed to implement data driven objectives as well as provide a ground up view of some of the hiccups and hurdles that we as implementers face when dealing with senior leaders.

Marc Voorhees Senior Manager Data Analyst Lead Pfizer

Why successful industrial analytics system needs automated AI/ML

Despite the breakneck pace of innovation, it isn’t often that we in the technology sector come across a technology as transformational as the Internet of Things (IoT). There are many research reports and real-world evidence indicating that IoT will transform our personal, public, and vocational experiences with efficiency gains and automated business processes, as well as optimized decisions. But, extracting these benefits require close interaction between three separate skill sets - the domain expert, the data scientist and the programmer. A typical business user rarely has all these three skills. Therefore, the only way to deliver the benefits of connected experience in most business and personal context is via Automated AI/ML which includes automating model selection, tuning, feature selection, data preparation, accuracy tracking and user knowledge adoption. For the domain user, the analytics system should be simple to setup, easy to understand to act on AI/ML based recommendations. In this talk, we will discuss the design of automated AI/ML in industrial applications with examples of Oracle IoT applications and discuss several customer use cases of how the automated system delivers anomaly detection, predictions and recommendations.

Viji Krishnamurthy Senior Director - Product Management Oracle

The next generation of Big Data platforms for advanced analytics - 100s of PetaBytes with real-time access

Building a reliable Big Data platform is extremely challenging when it has to store and serve 100s of PetaBytes of data in a real-time fashion. This talk reflects on the challenges faced and proposes architectural solutions to scale a Big Data Platform to ingest, store, and serve 100+ PB of data with minute level latency while efficiently utilizing the hardware and meeting the security needs.

In this talk, we'll dive into the technical aspects of how the ingestion platform can be re-architected to bring in 10+ trillion events/day at minute-level latency, how the storage platform can be scaled, and how the processing platform can be redesigned to efficiently serve millions of queries and jobs/day. We will provide a behind-the-scenes look at the current Big data technology landscape, including various existing open-source technologies (e.g. Hadoop, Spark, Hive, Presto, Kafka, Avro, Parquet) as well as what we had to build at Uber and open-source to fill the gaps and push the boundaries such as Hudi and Marmaray.

The audience will leave the talk with greater insight into how things work in an extensible modern Big Data platform and will be inspired to re-envision their own data platform to make it more generic and flexible for future new requirements.

Reza Shiftehfar Hadoop Platform, Engineering Manager Uber

Maximizing Talent: Getting the Most Out of Your Tech Teams

As AI continues to grow in popularity, the challenges of leading a tech team impacts nearly every industry, including Education, Finance, and Healthcare. Talent is not only hard to find, but leading a dynamic and complex group of experts and specialist can seem daunting. In this talk , we will unlock the keys to successfully leading a tech team from the perspective of great technical leader in science and technology.

Jennifer Shin Product Director NBC Universal

Chair closes

Main Stage: Chair's Opening Remarks

Alex Weber Award-winning host, motivational comedian speaker .

Main Stage Opening Keynote: Virtual Beings Not Virtual Assistants

Edward Saatchi CEO Fable Studio

Main Stage: Modern Day Technology

Dan Devone TV Personality NBC Sports

Building an online dating recommender system: lessons learned in practicality

One of the most relevant but overlooked challenges in adding machine learning capabilities to an existing product is to minimize disruptions to the tech stack and codebase. In this talk, Shanshan will describe how a small team at Hinge built a game-changing machine learning-based online dating recommender in a matter of weeks by solving this challenge. Moreover, she will discuss how they designed the system with maintainability and scalability in mind.

Shanshan Ding Director of Data Science Hinge

How Bloomberg Media uses Artificial Intelligence

Bloomberg.com consistently ranks among the top 10 most visited news and financial websites on the internet with around 60 million unique visitors a month. We use data to achieve a wide variety of business goals, including driving user engagement, increasing ad revenue, improving conversions for subscriptions, helping subscriber retention, and more.
In this talk, we will provide a broad overview of the different ways we use AI techniques and data to help us achieve these business goals. We will discuss how these techniques translate to our end-user experience, sometimes in very subtle (almost transparent) ways, and help move the needle. We will also present a case study where we used Machine Learning to personalize introductory offers to drive more subscriptions to our website, without cannibalizing revenue.

Dhaval Shah Director of Engineering Bloomberg

Understanding customers and crafting elegant experience for them

In an ever-changing marketplace and industry where new products are constantly emerging, data can play a big role in helping us learn and respond in a timely way to shifts in user interests, behaviors, and needs. Design can be informed by data. Design can also bring deeper meaning
to data. How might data experts work with User Experience Designers in the practice of designing experiments and collecting data, with the common goal of understanding customers and crafting elegant experience for them?

Peipei Wang User Experience Designer LinkedIn

Networking lunch

With innovative launchpad presentations from:
Nico Rode, TIBCO
Wade Tibke, Sigma Computing
Michael McNair, Codey.ai & 55B Labs

Spotting digital doppelgängers - behavioral clustering of users at scale

Be it for customer segmentation for ad targeting or understanding user behavior to devise product strategy, there is immense value in being able to group hundred of millions of users into a countable number of easily relatable user profiles / clusters. This talk will cover the various unsupervised machine learning techniques available for such a task along with the pros and cons of each technique in practice. In addition, we will cover ways of evaluating the quality of clusters produced by these unsupervised techniques

Jeffrey Tang Data Scientist - Team Investigations Tech Lead Twitter

Deep learning in online commerce

In this talk, we will discuss how recent advancements in artificial intelligence are transforming online retail. Deep learning-based image analysis and natural language processing open exciting new horizons for product discovery, ranking and recommendations. We will share our experience building practical and profitable solutions for online retailers using latest ML technologies and platforms.

Eugene Steinberg Technical Fellow and Principal Architect Grid Dynamics

Machine Learning and Artificial Intelligence at Marketing@Uber

Uber is part of the logistical fabric of more than 700 cities around the world. Whether it’s a ride, a sandwich, or a package, we use technology to give people what they want, when they want it.

Mario Vinasco Marketing Analytics Manager Uber

Models as a Service for real-time decisioning

This talk will lay down, step by step, the critical aspects of building a well-managed ML flow pipeline that requires validation, versioning, auditing, and model risk governance. We discuss the benefits of breaking the barriers of a monolithic ML use case by using a service-based approach consisting of features, models, and rules.

Join us to have an insight into the technology behind the scenes that accepts a raw serialized model built using popular libraries like H2O, SciKit Learn or Tensorflow, or even plain python source models and serve them via REST/gRPC which makes it easy for the models to integrate into business applications and services that need predictions

Niraj Tank Senior Manager, Software Engineering Capital One

Sumit Daryani Manager, Software Engineering Capital One

Chair's closing remarks

Stefan Byrd-Krueger Chief Analytics Officer ParsonsTKO

Main Stage: Chair's Opening Remarks

Alex Weber Award-winning host, motivational comedian speaker .

Main Stage Opening Keynote: Virtual Beings Not Virtual Assistants

Edward Saatchi CEO Fable Studio

Main Stage Panel: Spotlight on social media

- Best practice when dealing with large data sets
- How to upscale large ML models
- Understanding data ethics and privacy and how it might affect you in the future

Jeffrey Tang Data Scientist - Team Investigations Tech Lead Twitter

Peipei Wang User Experience Designer LinkedIn

Evren Eryurek PhD Director, Product Management Google

Main Stage: Modern Day Technology

Dan Devone TV Personality NBC Sports

Chairs opening remarks

Vasudha Badri-Paul CEO Avatara Digital

Marketing in 2020: Demystifying Smart Assistants & AI

• How voice-driven AI technologies are changing customer behaviour
• Preparing for the future of marketing with voice assistants like Siri, Alexa & Google
• Reimagining customer experiences like search, shopping, and CRM

Amy Bishop Strategy Director Epsilon

Propensity Scoring in CRM Audience Segmentation

A number of heuristic audience segmentation strategies exist in CRM marketing today. One of the most common of these is RFM (Recency, Frequency, Monetary Value), which works to identify segments that should be mailed more frequently based on the recency/frequency of engagements as well as total spend. Levi’s created an in-house audience scoring model building off derived rather than assumed relationships, which ultimately not only outperformed the incumbent RFM model in terms of audience activation, but established a flexible tool able to accommodate changing business priorities and established to incorporate new data sources as they become available. This talk will provide an overview of the ideation process, modeling approach taken, iteration/testing, and ultimate realization of the model and its business application.

Ben Saunders Senior Marketing Analyst, Ecommerce Levi Strauss & Co

Natural Language Use Cases at Uber

Franziska Bell, PhD Director of Data Science Uber

Lunch

With innovative launchpad presentations from:
Nico Rode, TIBCO
Wade Tibke, Sigma Computing
Michael McNair, Codey.ai & 55B Labs

Translating complex data into insights

Given that end-consumers work across leadership (VP+, GMs, C-suite and occasionally the Board), a big part of Paypal's job is to ensure that complex market data is synthesized into clear insights for varying audiences. Discover how to disseminate information through decks, readouts, dashboards and periodic “charts”.

Gaurab Bose Business Analyst PayPal

Getting personal with customers: Optimizing your omnichannel strategy using AI

• How AI impacts customer loyalty: Building consistency across touch points
• Effectively creating real-time customer experiences driven by data
• Using AI to nurture customer relationships both online and offline

Jason Mills Director of Product & Customer Experience Expensify

Deep learning 101: A lightning introduction - The Impact of AI on Visual Intelligence

Hive uses AI to solve the hardest questions for media companies: what content is most desirable and how do media companies prove the value of advertising spend? Hive's unique perspective as both a technology provider and data integration platform gives the company a lens into the most important ways that artificial intelligence will change the way we think about media content. In this talk, you will learn how deep learning has made its way into the media world, what it means for both content owners and advertisers, and how you can join the AI revolution.

Kevin Guo Co-founder & CEO Hive

Data driven product management: How to build a winning customer experience

• Utilizing AI to surprise and delight customers
• Gleaning insights to better understand customer journeys to develop products
• Building seamless experiences backed by AI

Cathy Tanimura Sr. Director of Analytics and Data Science Strava

Overcoming the data silo challenges: Solving the right problems through AI

• Not a one-size-fits-all approach: Finding answers in data sets big and small
• Using ML effectively to pick up patterns in user behavior
• Developing data literacy across the business to enhance capability and agility

Anna E Shen Marketing Manager, Paid Social Instacart

Chair’s closing remarks

Main Stage: Chair's Opening Remarks

Alex Weber Award-winning host, motivational comedian speaker .

Main Stage Opening Keynote: Virtual Beings Not Virtual Assistants

Edward Saatchi CEO Fable Studio

Main Stage Panel: Spotlight on social media

- Best practice when dealing with large data sets
- Understanding data ethics and privacy and how it might affect you in the future

Jeffrey Tang Data Scientist - Team Investigations Tech Lead Twitter

Peipei Wang User Experience Designer LinkedIn

Evren Eryurek PhD Director, Product Management Google

Main Stage: Modern Day Technology

Dan Devone TV Personality NBC Sports

How simple analysis can have a substantial impact on helping a company become more data-driven

Toca Boca is the the No. 1 mobile-first kids brand, creating digital toys and playful products for children around the world. The company recently conducted a seasonality analysis, which involved looking at revenue to determine optimal launch dates for products. Toca Boca's growth team used data to recommend target launch release dates for apps that maximized revenue potential. The result was record-breaking app revenue, as well as a complete shift in how Toca Boca produces apps - everything from forecasting financials to building timelines for game development teams. In her session, Nar will discuss what she learned from analyzing data over the course of a year, and how she used her knowledge to effect positive change for the company.

Nar Parisawan Senior Manager Analytics, Growth Lead Toca Boca

Succeeding in personalization at scale

The importance of personalization is repeatedly being shown to us across all industries, but none of these markets seem to have quite the potential for personalization as the gaming space does because we create an already dynamic, interactive product.
To unlock this level of potential, we need to develop not only the underlying tech that will execute this vision, but the techniques and algorithms that will allow insights to scale to such a level. In this talk, Scott will be illustrating methods that EA is using to build comprehensive understandings of their players’ behaviors to power cohesive messaging and recommendations across their numerous platforms and games. Specifically, the attempts to evolve from static moment-to-moment interventions to a holistic understanding of the player’s current state.

Scott Allen Director, Data Science Electronic Arts

Using AI analytics to detect real-time application issues, optimize gameplay experience, and maximize revenue

Bugs, imbalance, crashes, pricing glitches, and churn are all a day in the life of a gaming company. Gaming entities collect 100s of millions and in many cases billions of data points on a daily basis which makes traditional monitoring tools obsolete. Anodot has changed how gaming companies like King and Outfit7 detect, diagnose, and address abnormalities. Discover how a real-time AI powered analytics solution can accurately monitor, correlate, and alert when gameplay experience, and ultimately revenue, is negatively impacted by the maladies that constantly arise in the development and maintenance of a game.

Steven Kirkpatrick Solutions Engineer Anodot

Networking lunch

With innovative launchpad presentations from:
Nico Rode, TIBCO
Wade Tibke, Sigma Computing
Michael McNair, Codey.ai & 55B Labs

Hearthstone meta analysis

In Blizzard’s digital collectible card game, Hearthstone, players create 30-card decks and play them against one another, always keeping in mind what is currently top of the ‘meta'. Hear from Tian, Senior Data Scientist, on how they built a data pipeline from this raw gameplay data using unsupervised learning method (clustering) to then get deck prototypes from millions of deck content variations.

Tian Ding Senior Data Scientist Blizzard Entertainment

Preserving data while minimizing data footprint effectively for performance and costs

Raw event structures change over time and this impacts historical analysis of the dataset. Data Growth contributes to very large event tables which affects the performance of batch jobs and ad hoc queries. At GSN, Samson, Data Products Architect talks about how these issues are mitigated.

Samson Koshy Data Products Architect GSN Games

Round table discussion: Turning around a product with effective analytics

The product has launched. Everything seems to be going well but all of a sudden, it isn't. The metrics are going south, and the tricks that worked for a while are no longer working. This session, based on the speaker's own experience in the cut-throat mobile gaming industry, discusses how to quickly diagnose a situation like such and deliver timely solutions to turn around your product.

3 technology trends that will shape the next generation of gaming

The gaming industry has gone through dramatic changes in the last 10 years, but there are three trends that are likely to change things even further: the increased power of mobile gaming, the integration of AR in real-world environments, and the growth of streaming audiences eclipsing player growth.

In this session we’ll explore each of these movements and discuss how their convergence, supported by 5G connectivity, AI and Data, could shift the gaming business in new directions

Erik Archer Smith Head of ABM Treasure Data

Chair's closing remarks

Omeed Rameshni Director, Insights, League of Legends Riot Games

Chair opens

Alex Weber Award-winning host, motivational comedian speaker .

Round Table Discussion: Tapping Unstructured Data for Insights & Analytics: Business Imperatives & Success Factors

This roundtable will address the universal need to manage, integrate and leverage for business decision the torrents of unstructured data accumulating daily at the enterprise level. We’ll assess and discuss platforms, approaches, use cases, experiences, work arounds and new research and insights from deep learning techniques.

Joyanta Majee VP, Head of Data & Analytics Orion Business Innovation

When Can We Trust a Decision Made by a Machine: Building Trustable AI and Detecting Misinformation

With the advent of machine and deep learning, explainability and interpretability has become paramount to traceable and justifiable and explainable results:
when someone's house gets foreclosed, some one is sentenced, an insurance claim is denied, a mortgage application is denied. There are legal, human, organizational, IP and societal implications of leveraging the augmented intelligence of machines in supporting highly complex and previously only open to highly educated, highly skilled experts in medicine, underwriting, law, adjudication, etc.

The need to create and train unbiased or minimally biased datasets for deep learning in the interests of Fairness, Accessibility and Transparency requires best practices that are not known to organizations embarking on machine learning and AI activities.

In this session we will cover the methods and techniques and best practices to detect, manage and mitigate bias in datasets, deep neural network training, application integration. We will also explore the cognitive biases that lead us to curate date to create less than trustworthy blackbox AI systems, and how to avoid them.

We will explore in detail (code level as well as architecture), the project AlternusVera which detects fakeness or deliberate misinformation in a body of text.

Ali Arsanjani Vice President, Artificial Intelligence Deep Context

Magic Dust for Artificial Intelligence Product Management; specific skills, techniques, attitudes and responsibilities for PMs when it comes to AI-driven products.

Explore particular challenges PMs face when dealing with AI, like setting user expectations, working with messy data and embracing probabilistic and imperfect results.

Mark Cramer Applied Artificial Intelligence Product Management Xerox

How Data and Machine Learning enable Creatives and Storyteller

Stefano Corazza, PhD Sr Director, Engineering - Head of AR Adobe

A New Ecosystem Approach to Improve Data Science Success

- The dynamic nature of deep learning methods
- Creating greater personalisation of customer analytics
- Improving accuracy and performance in applications

Dr Alex Liu Chief Data Scientist IBM

Creating Iconic Viral Trends: A Look Into the $500m + Fidget Spinner Phenomenon

Creating a marketing campaign that “goes viral” is the goal of every marketer. With product marketing, it makes that goal even more satiable. However, the approach to product marketing goes much deeper than the sales enablement and marketing campaigns. The key doesn't lie in pay-to-post Facebook ads, algorithms and various other marketing nuances geared to track SEO numbers and customer engagement. Most companies don't understand how much that hurts their marketing. With in-your-face sales ads now being a thing of the past, consumers just want more. The secret to creating a campaign that is going to 'go viral?', isn't accidental or methodical, viral trends are about depth. Maneuver your product positioning, emotionally connect with your target consumer through content marketing and events, to effectively build your community.

Rob Fajardo CEO Leave Normal Behind

Experimentation @ DoorDash

Lessons learned about how to run experiments and make informed business decisions leveraging data.

Jessica Lachs VP of Analytics DoorDash

Lunch

With innovative launchpad presentations from:
Danielle Deibler, Marvelous.ai
Ted Benson, Instabase
Pat Giblin, 451 degrees

Building the Cities of the Future: Smart Cities, Startups and Minimum Viable Policies

Thoughtful integration of city planning, smart cities technology and policy development that puts people first is critical for governments and societies to adapt and thrive in an AI-powered world. As major cities around the globe continue to experience population growth, city governments will face increasing challenges to meet the needs of its communities. Advances in artificial intelligence are expected to replace millions of jobs and breakthrough technologies are being deployed faster than public policies, posing both incredible opportunities and unanticipated consequences on society. Breaking data silos across city departments and municipalities, cultivating a data-driven culture and creating a data-skilled workforce within government, as well as building strategic partnerships with the Govtech community, are key to successfully leveraging the full potential of data analytics to help to build a better foundation for the cities of the future.

Michelle Littlefield Program Lead, Digital Services & Analytics City of Redwood City

Data Strategies for Enterprise: Model Risk; Operationalisation of Data Insight

- Innovations in the industry
- Dealing with large amounts of data
- Understanding the advanced database platform

Omeed Rameshni Director, Insights, League of Legends Riot Games

Chair's opening remarks

Stefan Byrd-Krueger Chief Analytics Officer ParsonsTKO

Enable machine learning for everyone - An ML case study for abandoned online cart data

Recent development on the machine learning packages and DevOps platforms makes it possible of mass participation for maximum business value. In this talk, we will review those developments and present a case study (Abandoned Cart) as part of T-Mobile’s effort to scale up Machine Learning along with retrospective thoughts on lessons and experiences.

Yunhang Chen Principal Engineer T-Mobile

Artificial Intelligence: Building the Future of E-Commerce

In this talk, we will explore the applications of different facets of AI (machine learning, deep learning, NLP and computer vision) in online retail and e-commerce. We will walk through practical examples of how AI can be used to personalize customer experience and revolutionize your customer journey through personalized marketing campaigns, personalized recommendations, supply chain, CRM, reranked search and more. Kamelia will share her thoughts on building successful AI teams and an experiment-driven culture that will reshape your business by enhancing customer experience through AI.

Kamelia Aryafar Chief Algorithms Officer Overstock.com

Computer vision from space

Planet operates the largest constellation of satellites in human history, capturing 1.5 million images of Earth on a daily basis. These images are funneled into their analytic engine, serving intelligence products that are revolutionizing how many industries measure global activity. In this talk, Jesus will present how raw pixels from space are transformed into API feeds of analytic information. He will describe some specific applications, like deforestation monitoring, ship detection and mapping of building and road footprints all over the world. In addition, he will share some lessons learned in the development of machine learning models and datasets.

Jesus Martinez-Manso Engineering Manager, Machine Learning Planet

Machine Leaning in Education

This session will cover key technical challenges in processing academic documents and how they differ from general purpose NLP techniques applied to solve non-structured text use-cases.
In addition we will discuss various Machine Learning solutions to augment academic documents with rich meta data information extracted from these documents.

Johnson D'Souza Senior Machine Learning Engineer Course Hero

Networking lunch

With innovative launchpad presentations from:
Danielle Deibler, Marvelous.ai
Ted Benson, Instabase
Pat Giblin, 451 degrees

Experimentation-driven customer experience

Kuntal offers use cases, lessons learned and business impacts made using customer experience data and experimentation. She shares how combining quantitative data with unstructured quantitative data could provide powerful insights into user behavior that can be leveraged for improving business KPIs. Her team is using AI to support thousands of employees at PayPal and fueling innovation and data driven culture

Kuntal Maya Goradia Director of Analytics PayPal

Human in the loop Machine Learning

Fraud is an adversarial problem where attackers are constantly working against the system to identify and exploit any loopholes that might be present in the system. Unlike most machine learning applications where labels can be automatically inferred by the system, the underlying signatures of fraud attacks are often novel and varied. This makes it critical for humans to work together with the machines and enable the human in the loop system to better discover blind spots and identify novel fraud attacks. In this talk with Affirm’s Head of Data Science, Nitesh Kumar will explain the human-in-the-loop approach and how this model can fundamentally drive machine learning based fraud systems to be more efficient. Additionally, he will discuss how machines can enable and guide human intervention through active learning.

Nitesh Kumar Head of Data Science Affirm

Main stage: DATAx Start-up Showcase

Hear from the most innovative start-ups in San Francisco and how they will change the data landscape in the next few years.

Alex Weber Award-winning host, motivational comedian speaker .

Chair's opening remarks

Vasudha Badri-Paul CEO Avatara Digital

Approaches to partnering for data sciences in healthcare

Data sciences innovation is flourishing around the world, but how do we harness that innovation to transform the healthcare industry? How do entrepreneurs, academics, and the healthcare industry work together to help take products and technologies from concept to commercialization, and ultimately, impact the lives of millions worldwide? Both industry and entrepreneurs benefit from partnerships, but they can often be challenging to build and maintain, particularly in novel technology areas such as AI which have not yet fully established trust in their value. At Johnson & Johnson Innovation, we build relationships and create customized collaborations with regional entrepreneurs, universities and institutes developing early- to mid-stage innovations across Pharmaceutical, Medical Device and Consumer Health sectors. I will discuss some of the unique challenges related to data sciences and AI on both sides of the partnering equation, and suggest approaches which facilitate partnering to accelerate the impact of this technology in healthcare.

Emma Huang Director, Data Sciences, External Innovation Johnson & Johnson

Promises & pitfalls: Using ‘big’ medical records data for research

The widespread adoption of electronic health records has facilitated passive collection of large amounts of computerized medical data. Researchers are eager to leverage these data into insights that can meaningfully improve clinical care and patient outcomes. Yet enthusiasm about the impressive size and availability of these datasets should not diminish our awareness of their weaknesses; as with all research, it is essential to draw careful conclusions that are well supported by the data. This talk will give an overview of electronic health record data, review its potential strengths, and outline five common pitfalls, with recommendations on how to mitigate them.

Kathryn Rough Research Scientist Google

Driving AI with Human-Centered Design

The Innovation Force at Cambia Health Solutions will walk through a case study of how they recently used a human-centered design process to drive development of better products using AI and ML. By designing to solve customer problems, they developed a new way to give peace of mind to consumers paying medical bills.

Max Janasik Vice President of Innovation Cambia Health Solutions

Nicole Cathcart Director, Innovation Cambia Health Solutions

Segmentation of users by health condition and prediction of future diagnoses

Expert systems have long been used in healthcare to classify people into segments based on their medical conditions. However, these approaches can be very brittle, as well as expensive to build and maintain, when used to classify people who are in the early stages of a condition or when the available data is sparse or messy. Machine learning and, in particular, deep learning, can often outperform an expert systems approach in these scenarios in accuracy and cost. This talk will cover the design and results of a system Castlight built that combines deep learning, traditional machine learning, and expert systems to classify users into segments related to their medical conditions using their medical and drug claims, demographic data, application activity, and biometrics. We use the resulting segmentation to deliver personalized recommendations to users directing them to relevant employer-sponsored benefit programs, educational content, and high quality, low-cost medical providers for their condition.

Robert Stewart CTO & Chief Architect Castlight Health

The rise of informatics - A sepsis use case

Healthcare has been transforming from a volume model to value, where patient satisfaction is at the forefront. Data driven decisions and timely information are now becoming a critical component to create value by making patients and consumers the core of their strategies. Data driven approaches can help improve preventative care, drive business decisions, and reduce overall costs of care. This talk will go over why healthcare needs big data and how healthcare institutions like Sutter are using data, specifically Informatics to improve patient care. This talk will address a data driven case on Sepsis and the direction Informatics is heading.

Kriti Bhatia Business Intelligence Analyst Sutter Health

Leveraging Distributed Systems & AI for Multi-Omics Data Analysis

Henry Ines CEO Shivom

Lunch

With innovative launchpad presentations from:
Danielle Deibler, Marvelous.ai
Ted Benson, Instabase
Pat Giblin, 451 degrees

Optimizing marketing data to personalize healthcare

Beachbody LLC is the creator of the nation’s most popular fitness and weight-loss solutions including P90X, Insanity, 21 Day Fix. Customer focus is key for us, and the direct-selling division of Beachbody is powered by our independent distributors, who are called as “Coaches”. Data is critical at Beachbody for making important decisions. In this session, the speaker, Aarthi Sridharan will be doing a deep dive on how Beachbody leveraged its data platforms to build an in-house email marketing system to help marketing team to customize and personalize communication to customer and coaches. In addition to saving millions of dollars, this solution enabled us to better segment & target customer by collecting customer, order transactions, social media, web analytics and email response data. This personalization of emails and communication resulted in better click rate & customer conversion and open rate increased by 10%.

Aarthi Sridharan Senior Director, Data Beachbody

Mobile device data in healthcare

Recent technological advancements make it possible to closely and continuously monitor individuals on multiple scales in real time while also incorporating genetic, environmental, and lifestyle information. We are collecting and using this multi-scale biomedical data to gain a more precise understanding of health and disease at molecular and physiological levels and developing actionable, predictive health models for improving cardiometabolic outcomes. We are simultaneously developing tools for the digital health community, including the Digital Biomarker Discovery Pipeline (DBDP), to facilitate the use of mobile device data in healthcare.

Dr Jessilyn Dunn Assistant Professor, Departments of Biomedical Engineering and Biostatistics & Bioinformatics Duke University

Chair's opening remarks

Alan Burke Director of Analytics Wizards of the Coast

Behavioral economics & game monetization

A number of lessons from the field of behavioral economics can inform game monetization. This presentation will cover a few concepts, run through the experiments in these areas, and explore the implications for increasing game monetization (emphasis on mobile games).

Kian Sandjideh Former Director of Online Operations Electronic Arts

Working cross-functionally to bridge the gap between creative and analytical teams

Analytics is a discipline that relies on managing constraints, while creativity is – by design – an exercise in avoiding constraints. Explore in this session how WB Games has developed a discipline that enables developers, marketers and analysts to work at the speed of creativity.

Matt Howell Executive Director, Analytics Warner Bros. Games

Scaling a data science team

Throughout the games industry, the perennial challenge of scaling data science teams remains key to an organization’s success. Understanding how to balance specialization and diversification helps identify how to best build a cohesive data team.Join Florent as he brings his experience from Ubisoft in outlining how to get the most value out of your data science team and ultimately how to successfully be a part of and create the ideal gaming analytics team.

Florent Blachot Associate Director, Data Science Ubisoft

Data Strategies for Enterprise

Omeed Rameshni Director, Insights, League of Legends Riot Games

Measuring what matters: identifying the right metrics

In the world of content analytics at Twitch, many business metrics are created to measure content performances. Good metrics are effective in answering business questions, while bad metrics create confusion. It is crucial to identify the right metrics that align with the decision making process. This session explores the good and bad metrics in understanding different content performances.

Sharon Lin Analytics Manager, Content Twitch

Networking lunch

With innovative launchpad presentations from:
Danielle Deibler, Marvelous.ai
Ted Benson, Instabase
Pat Giblin, 451 degrees

Engaging with your Product Managers

Working alongside Product Managers is a key relationship to understand and manage as an analytics professional. Join this session see how Zynga ensure their analytics teams build the tools and technology to help product teams move faster and more efficiently.

Ryan Glosson Lead Product Manager Zynga

Main stage: DATAx Start-up Showcase

Hear from the most innovative start-ups in San Francisco and how they will change the data landscape in the next few years.