WeAreDevelopers Logo WeAreDevelopers
Sign in
For You LIVE
World Congress 2024 • Aug 20, 2024 • Session details

Building the platform for providing ML predictions based on real-time player activity

Artem Volk , Fabian Zillgens

How do you process millions of telemetry events to personalize gameplay in under 500 milliseconds? Explore Phoenix Games' decoupled, fault-tolerant machine learning architecture on AWS.

Pause
Mute Enter Fullscreen
#1 about 2 min

Overview of Phoenix Games and real-time customization goals

The challenge of adapting player experiences in real-time requires fast data iterations and automated process pipelines.

#2 about 2 min

Customizing player experiences in the Emergency HQ title

Machine learning models apply dynamic gameplay modifications to surface relevant store purchases alongside tailored mission difficulties.

#3 about 2 min

Microservice architecture requirements for unobtrusive event collection

Delivering dynamic game features demands simple cross-studio integration without heavy client SDK dependencies.

#4 about 2 min

AWS infrastructure stack and data flow pipeline overview

A simplified architecture highlights the event collection, offline data processing, and machine learning customization mechanisms.

#5 about 3 min

Managing raw game event execution and payload structures

Handling millions of out-of-order game stream instances requires distinct tracking of real event occurrence timestamps versus upload times.

#6 about 3 min

Comparing offline data analytics with online stream processing

Utilizing Apache Spark extracts long-term insights while real-time user activity is handled via unbound Apache Flink queries.

#7 about 2 min

Managing fast real-time player states using profile APIs

Storing live active user snapshot profiles enables rapid integration of both historical offline data and instant inferences.

#8 about 2 min

Demonstrating pipeline latency thresholds during player purchases

Tracking a native player transaction through Flink visualizes how the profile snapshot API updates within hundreds of milliseconds.

#9 about 2 min

Delivering dynamic game packages using external decision models

Decoupled AWS lambda functions empower independent machine learning calculations and simple rule-based feature testing configurations.

#10 about 2 min

Analyzing customization payloads alongside model inference decisions

Displaying specific package deals relies on observing distinct machine learning delays and debugging underlying player delay logic.

#11 about 2 min

Enabling autonomous deployments and independent monitoring controls

Empowering data scientists with Terraform dashboards minimizes blockages while securing the pipeline against flawed algorithm updates.

#12 about 1 min

Measuring API infrastructure limits and platform pipeline latency

Analyzing sub-second turnaround constraints uncovers operational bottlenecks when calling endpoints and executing live updates simultaneously.

#13 about 5 min

Evaluating deployment pipeline compromises and infrastructure scaling costs

Balancing the advantages of resilient decoupled microservices highlights fundamental trade-offs surrounding duplicate data and high storage costs.

#14 about 2 min

Processing extreme event volumes and validating machine operations

Operating at massive scale emphasizes the importance of data verification to prevent custom features from hurting core business revenue.

#15 about 3 min

Executing automated model generation and dynamic user packaging

Empowering data analysts to build localized cloud containers directly enables specific item delivery configurations tailored to individual gamers.

#16 about 4 min

Implementing Lambda request load balancing and player privacy controls

Handling high volume interactions requires strict AWS container load provisioning and limited personal asset storage tracking parameters.

Matching moments

4:29 min

Shifting data analytics from monetization to player experience design

Johanna Pirker Johanna Pirker · World Congress 2021

2:13 min

Modernizing legacy applications for real-time streaming data consumption

Farooq Sheikh Farooq Sheikh +3 · World Congress 2025

4:31 min

Optimizing system scalability and real-time push capabilities

David Leitner · World Congress 2022

2:59 min

Integrating data analytics and machine learning as interactive elements

Johanna Pirker Johanna Pirker · World Congress 2021

2:56 min

Scaling experimental game concepts into production fitness products

Daniel Meilak Daniel Meilak +1 · World Congress 2026 Europe

5:44 min

Behavioral profiling and creating custom player types in games

Johanna Pirker Johanna Pirker · World Congress 2021

More from World Congress 2024

Beyond the Hype: Real-World AI Strategies Panel thumbnail
24:07
12 months ago

Beyond the Hype: Real-World AI Strategies Panel

Mike Butcher, Jürgen Müller, Katrin Lehmann, Tobias Regenfuss

Carl Lapierre - Exploring Advanced Patterns in Retrieval-Augmented Generation thumbnail
21:17
almost 2 years ago

Carl Lapierre - Exploring Advanced Patterns in Retrieval-Augmented Generation

Carl Lapierre

Coding an Immersive Copilot using Unity / .NET and Azure OpenAI! thumbnail
31:50
about 2 years ago

Coding an Immersive Copilot using Unity / .NET and Azure OpenAI!

Zaid Zaim

Building a diverse tech company: The power of Diversity, Equity, Inclusion, and Belonging thumbnail
29:43
about 2 years ago

Building a diverse tech company: The power of Diversity, Equity, Inclusion, and Belonging

Sue Quackenbush

Dapr in Practice thumbnail
24:08
about 2 years ago

Dapr in Practice

marcklefter

Minimising the Carbon Footprint of Workloads thumbnail
27:46
about 2 years ago

Minimising the Carbon Footprint of Workloads

Michael Mueller

Upcoming sessions on this topic

Open session

World Congress 2026 North America

September 25, 2026 · 09:00–09:30

Stage 7

AI Agents are Only as Smart as their Context: Building a Real-Time Context Engine

Bharat Patel

Lead Software Engineer at Intuit

Bharat Patel
Open session

World Congress 2026 North America

September 24, 2026 · 14:10–14:40

Stage 3

Real-Time Data Platforms at Trillion-Event Scale

Diptamay Sanyal

Principal Engineer | Data, AI & Cybersecurity Platforms

Diptamay Sanyal
Open session

World Congress 2026 North America

September 23, 2026 · 10:45–12:45

Stage 10

Agents That Own Their Inference: Building Production AI Agents on Dedicated GPUs

Khaja Omer, Sheilah Kirui

Khaja Omer
Sheilah Kirui
Open session

World Congress 2026 North America

September 25, 2026 · 16:50–17:20

Stage 7

You Can’t Rerun Sunlight: Designing ML Data Architectures for Physical AI

An Phan

Senior Data Infrastructure Engineer @ Hippo Harvest

An Phan
Open session

World Congress 2026 North America

September 24, 2026 · 17:30–18:00

Stage 3

Scaling Distributed Queues for AI workloads

Jasmit Kaur Saluja

Staff Software Engineer at Meta Platforms Inc

Jasmit Kaur Saluja
Open session

World Congress 2026 North America

September 24, 2026 · 14:10–14:40

Stage 7

Designing High-Performance AI APIs: Lessons from Serving Millions of Real-Time Requests

Wayne Liu

Chief Growth Officer and Americas President of Perfect Corp.

Wayne Liu

Related videos

How Data is Shaping our Games thumbnail
43:30
World Congress 2021 • about 5 years ago

How Data is Shaping our Games

Johanna Pirker

Machine Learning for Software Developers (and Knitters) thumbnail
49:45
LIVE • over 5 years ago

Machine Learning for Software Developers (and Knitters)

Kris Howard

From 0 to 1.000.000: How to build a serverless raffle service for hyperscale thumbnail
30:03
World Congress 2023 • almost 3 years ago

From 0 to 1.000.000: How to build a serverless raffle service for hyperscale

Marco Plaul, Martin Sakowski

Remote Driving on Plant Grounds with State-of-the-Art Cloud Technologies thumbnail
48:54
LIVE • about 5 years ago

Remote Driving on Plant Grounds with State-of-the-Art Cloud Technologies

Oliver Zimmert

Tracking vehicles at scale thumbnail
30:33
Europe 2026 Virtual • 3 months ago

Tracking vehicles at scale

Thorsten Riess

Swapping a Data Warehouse at Runtime: Zero-Downtime Migration Without Changing a Single Client thumbnail
30:28
World Congress 2026 Europe • 3 months ago

Swapping a Data Warehouse at Runtime: Zero-Downtime Migration Without Changing a Single Client

Max Fischer, Michael O'Toole

Related articles

From learning to earning

Jobs that call for the skills explored in this talk.