Data Con LA 2019
| title | author | length |
|---|---|---|
| The challenges of data science for veteran media organizations | Stan Dyro | 46m |
Small team modernizes a legacy newspaper into a tech product by rebuilding pipelines, warehouses and tooling rather than chasing hype. Core stack centers on S3… | ||
| Box Office Strategy: Do Release Dates Matter? | Preet Singh | 43m |
Uses an unofficial GroupLens/Kaggle 45,500 movies dataset and revenue/budget fields to test whether Memorial Day sinks box office. Finds Friday is only ~30% of… | ||
| Large scale streaming analytics | Jay Palaniappan | 39m |
Shows an AWS serverless reference architecture that actually ingests, stores, analyzes and monitors streaming data: 100K records/second is the stated target… | ||
| Applied Privacy Engineering Study | Ken Chang, Serena Villalobos | 40m |
Analysis of the SEER cancer registry (1991–2000, ~300k records) shows many records have k = 1 under quasi‑identifiers, meaning direct reidentification via… | ||
| Working with data from your most engaged customers | Jeff Cooper | 41m |
FabFitFun turns its engaged forums and surveys into product-level signals: 100+ predictive surveys a season and post‑box product feedback feed per-user ratings… | ||
| MetaConfig driven FeatureStore | Piyush Kumar | 51m |
MakeMyTrip - India's #1 online travel platform having more than 70% of the traffic from mobile apps embarked on a journey to revolutionize its customer… | ||
| Data warehouse and Kubernetes: Lessons from ClickHouse Operator | Robert Hodges | 41m |
An operator that deploys ClickHouse on Kubernetes actually provisions StatefulSets, persistent volume claims, and a Zookeeper ensemble from a single custom… | ||
| Scrum Overview | Sonia Macias | 35m |
Case study of switching to one-week sprints that moves a team from ~11–13 story points to 32.5 story points over several sprints, delivering Tableau dashboard… | ||
| Location Analytics For Smart Grid Reliability | Vivian Sultan | 37m |
Research shows location analytics pinpoint where interventions cut outages and risk: a spatially enhanced analytical model ties OMS/SCADA/asset records to… | ||
| User Motivation: Refining Customer Segments with Location | Annie Flippo | 29m |
Combines deterministic SDK location logs with demographic and offline interest data to build industry-specific segments using k-means clustering and… | ||
| Move Fast, Think Big | Boaz Brudner | 50m |
Focus on delivering only what actually lands in production, using quarterly plans plus OKRs over ad hoc sprint churn, and cut work that repeatedly stalls or… | ||
| Know Your Customer | Fiona Li | 45m |
Ticketmaster implements RFM and probabilistic models (Bayesian, Beta-Geometric/Beta-Binomial, gamma for spend) to predict two-year LTV at the individual level… | ||
| Big Data Modeling with Spark SQL: Make data valuable | Jayesh Patel | 36m |
Shows how to build a single wide provider analytics fact for 80,000 providers (100M patients) using Spark SQL and Kudu to avoid joining large fact tables, cut… | ||
| E-Commerce Product Categorization at Scale | Jesse Steinweg Woods | 38m |
Honey builds an ensemble that combines a text CNN (title + merchant tokens), an image transfer model (Inception backbone), Airflow orchestration and model… | ||
| Unifying streaming and message queue with Apache Kafka | Mingmin Chen | 43m |
Uber builds a message-queue layer on Apache Kafka that adds dead letter queues, ack/nack out-of-order with configurable retries and delay, and in-process… | ||
| Real Time Face Detection through a Webcam Using AI | Obaid Sarvana | 49m |
Builds a working real‑time face detector that runs entirely in the browser using TensorFlow.js and face-api.js with a webcam stream, WebGL acceleration, and… | ||
| HTAP with MariaDB Platform | Todd Stoffel | 41m |
MariaDB Platform combines the MariaDB server, the MaxScale proxy and the ColumnStore engine to run HTAP on one entry point, with MaxScale doing routing… | ||
| Pitney Bowes methodologies | Dan Kernaghan, Tim Kreytak | 42m |
Pitney Bowes and Ironside describe a production pipeline: preprocess billions of spatial rows into a spatial data lake, join to 1,400 datasets via a persistent… | ||
| Don't Talk Data to Me | Jacklyn Agor | 38m |
Argues that data should stay in the appendix and decisions sell as human benefits: email yields 40:1 email ROI, Coachella-style app messages drive on-site… | ||
| Time Series Forecasting | Jatinder Singh, Rashmina Menon | 35m |
GumGum processes ~30 billion ad requests and 25+ TB of raw auction data per day and builds a real-time inventory forecasting pipeline that returns results… | ||
| Careers in Data Science | John Peach | 44m |
Breaks data careers into concrete job families (business analyst, BI engineer, data engineer, ML engineer, economist, research and ML specialists) and shows… | ||
| Data Science Education. Building Knowledge Graphs | Jose Luis Ambite | 41m |
Builds a public index that currently contains 12,000 resources (mostly videos) by scraping MOOC platforms, YouTube and lecture sites, extracting transcripts… | ||
| Open Data & Maps As Tools for More Impact | Juan Vasquez | 42m |
City of Los Angeles turns open datasets into operational tools using ArcGIS Online and free ESRI web app templates embedded in Drupal to replace static lists… | ||
| Data Acquisition Made Easy | Kaylin Linke | 43m |
Socialgist supplies an enterprise pipeline for Reddit with three concrete delivery options: an HTTP full feed of every post and comment (20M monthly posts… | ||
| Exploratory Data Analysis with Spark and IntelliJ IDEA | Maria Khalusova | 35m |
Demonstrates exploratory analysis on an IntelliJ IDEA code dump using Spark and an IntelliJ plugin that runs Zeppelin-style notebooks inside IDEA, converting… | ||
| Orchestration of Blue-Green deployment model | Chaitanya Rudrabhatla | 42m |
Demo shows a concrete pipeline that provisions an AWS VPC with public/private subnets and security groups, creates a DocumentDB cluster from a snapshot, and… | ||
| Integrating Kafka with a Real-Time Database | David Anderson | 41m |
Live demo shows MemSQL pulling Kafka topics into a real-time database on AWS, ingesting around 300k–1M rows/sec on a tiny two-node cluster while storing Kafka… | ||
| So You got Hacked, how Quickly Can your Company Recover? | Don Malloy | 42m |
Companies detect breaches after about 214 days and take months more to contain them; Atlanta pays $52k ransom demand and ends up spending ~$17M to recover. IoT… | ||
| Securing IoT Data with Pervasive Encryption | Eysha Shirrine Powers | 50m |
13 billion breached records since 2013 and only 4 percent encrypted makes the case concrete: protect IoT health data at every hop — device, network, app… | ||
| Data is the New Business Currency | Jim Benedetto | 44m |
Claims that the datasphere (33 ZB in 2018 to 175 ZB by 2025, 80% unstructured) demands a new pipeline: collect → process → store enabled by four things —… | ||
| Electrifying Big Issues Through Analytics | Mukund Kaushik | 38m |
Southern California Edison uses its sensor and enterprise data across 15 million customers, 1.4 million poles and 1 million tracked trees to run daily… | ||
| One (Key) Ring to Rule Them All | Shweta Shetty | 33m |
Teradata Vantage integrates Keycloak as the identity provider to deliver federated IAM and single-sign-on across Vantage web UI, Jupyter/Studio tools and the… | ||
| How to Teach an Old Firm New Tricks | Amarita Natt | 44m |
Explains concrete tactics for migrating legacy analytics to R, Python, SQL and AWS while keeping SAS and Excel users productive: build internal boosters across… | ||
| Graph Theory Takes on the Opioid Epidemic | Mark Quinsland | 42m |
Uses Neo4j graph databases, Cypher queries and Jupyter Python notebooks to link prescriptions, providers and pharmacies and detect fraud rings; Neo4j runs… | ||
| Patterns for Persistence and Streaming in Cloud Architectures | Jeff Carpenter | 47m |
Shows concrete patterns using Apache Cassandra and Apache Kafka: a live nine-node cluster spread across three data centers (AWS, GCP, Azure) replicates data… | ||
| Innovating with Data | Joey Beidatshoo | 45m |
Three business models are defined: data as a competitive advantage, data as a product, or data that merely improves existing offerings; companies like GE buy… | ||
| Best Practices for Prototyping ML Models for Healthcare | Lorenzo Rossi | 39m |
Practical checklist for prototyping ML on EHR: define cohorts by prediction time and discard future data to avoid temporal leakage, split train/test by time… | ||
| Optimization Algorithms for Deep Learning | Ash Pahwa | 41m |
Covers why plain gradient descent stalls on flat or tiny-gradient regions and how practical fixes speed training: momentum + Nesterov to push through plateaus… | ||
| State of the Art of Innovation in Computer Vision | Christian Siagian | 42m |
Survey of CVPR 2018–2019 highlights concrete shifts: the ImageNet/AlexNet jump (20% error improvement) enabled end-to-end convolutional pipelines running on… | ||
| nothing matches | ||