AWS Summit Sydney 2025
2025-06-04..2025-06-05
Sydney, Australia
| title | author | length |
|---|---|---|
| Education revolutionised with generative AI | Kevin Bell, Michael Blumenstein, Bert Verhoeven | 27m |
Universities build a secure enterprise AI stack, deploy a custom tutoring agent called Q reframer that reframes student questions rather than giving answers… | ||
| A day in the life of a serverless developer | Seb Kasprzak | 31m |
A developer turns a synchronous contract API into an async pipeline the same day, replacing a distributed transaction with API Gateway → SQS native… | ||
| Accelerate VMware migrations using AWS Transform | Sheng Chen, Senay Swinney | 27m |
AWS Transform automates VMware-to-Amazon EC2 migration using four LLM-driven agents: the discovery and data ingestion agent pulls vCenter inventory via RVTools… | ||
| Kubernetes workloads with Karpenter & Amazon EKS | Frank Fan, Faisal Masood | 31m |
Shows EKS Auto Mode plus Karpenter provisioning nodes on demand, scale from zero to serve a pod, then scale to larger instance types and consolidate by… | ||
| Zero-Trust Application Networking with VPC Lattice | Mahmoud Ismail | 27m |
VPC Lattice promises zero‑trust application networking that removes routing tables and gives each service a DNS entry and centralized control. It builds… | ||
| Different Customer Experience with Amazon Connect | Douglas Park | 29m |
Amazon Connect runs the same contact-center tech behind 150,000 agents and now adds real‑time transcription, sentiment analysis and instant subsecond summaries… | ||
| Elevating generative AI and LLM observability | Adam Gardner, Pankaj Bishnoi | 29m |
Generative AI adoption requires end-to-end observability: Amazon Bedrock for models and guardrails, Dynatrace AI/LLM Observability for token, cost and trace… | ||
| How AWS accelerates business transformation | Nicole Ong, Priyank Pashine | 29m |
AWS shows a migration playbook with measurable outcomes: AWS Transform uses agentic generative AI to discover, plan, convert code and produce IaC, automating… | ||
| Build and scale reliable multi-agent systems on AWS | Melanie Li, James Luo | 29m |
Built a production multi-agent system using Amazon Bedrock foundation models with a Bedrock Agents multi‑agent pattern, a web search tool hitting the ATO site… | ||
| Keynote: Builders Day | Rianne van Veldhuizen, Dave Brown, Jessie Hughes, Jonathon Creenaune | 01h 05m |
Keynote outlines concrete AWS wins and tools: Amazon Bedrock for managed foundation models, end-to-end RAG and model distillation that claims up to 500% faster… | ||
| Power of Possibility: Navigating the tech landscape | Ambika Sinha, Janette Tang | 01h 07m |
AWS helps Swimming Australia since 2018 build a data lake foundation that ingests competition, training and federation race history, then deploys computer… | ||
| Westpac's AI innovations for customer experience | David Walker, Jamie Simon | 32m |
Westpac describes an internal AI Accelerator that evaluates over 1,000 use cases and standardizes them into four patterns (information retrieval, process… | ||
| Building a culture of security with AWS and CBA | Mitch Beaumont, Harvey Deak | 27m |
Claims a distributed ownership mechanism called Security Guardians or a Security Champions program speeds delivery and reduces defects: AWS reports teams with… | ||
| Application modernisation with generative AI | Bilal Alam | 28m |
A customer prototypes a combined pipeline using Amazon Q Developer for bulk code transformation and Amazon Bedrock for prompt-driven fixes and test generation… | ||
| CommBank supercharged AWS Well-Architected reviews | Rovan Omar, Yuri Belenky | 29m |
Commonwealth Bank integrates Amazon Bedrock with an internal pipeline and the DevOps Hosting Platform (DHP) to auto-map CloudFormation/Terraform/diagrams plus… | ||
| Usage based cost optimisation at scale | Joshua Samuel, Chaitanya Krant, Julian Pittas | 27m |
National Australia Bank builds a data-driven power management platform that uses VPC flow logs ingested via Data Firehose into S3, an IP-to-application map… | ||
| How nib transformed customer experience | Milly McCutcheon, Mathew Finch | 21m |
nib uses Amazon Bedrock and LLMs on top of Amazon Connect, Lex and Textract to power a conversational agent Nibby across chat and voice (4+ million… | ||
| Modernising banking with Judo Bank | Kinnera Vajjala, Lawrence Hunt | 28m |
Judo Bank migrates its core to Thought Machine on AWS using serverless Lambdas, Confluent Kafka, Aurora, Terraform and New Relic, moving 8,000 lending… | ||
| Get started with Amazon Aurora DSQL | Peter Vandaele, Surendar Munimohan | 28m |
Amazon Aurora DSQL is a Postgres-based serverless relational database that autodiscovers and partitions storage, scaling from Aurora’s 128 TB limit toward… | ||
| Master large-scale ML batch inferencing | Warren Wei, Brett Tully | 30m |
Nearmap runs a pipeline-of-pipelines on 10,000-node EKS cluster using Kubeflow, Kafka, RDS and FastAPI to process imagery at scale, storing 100 PB in S3 and… | ||
| Building Your Intelligent Data Agent | Surya Kakanadan, Jan Richter | 24m |
Demo shows an agent that translates natural language to SQL using Amazon Bedrock Structured Knowledge Bases then runs queries against Amazon Redshift and Data… | ||
| How a retailer modernised their SAP landscape | David Nye | 18m |
A large Australian retailer merges two brands’ finance backends — a 6 TB ECC system and a 1 TB S/4HANA — into a single S/4 under SAP Rise on AWS in eight… | ||
| Worley's next gen data intelligence platform | Emily Barretto, Samit Sharma | 26m |
Worley moves from monoliths to a fully serverless multi-regional platform built on S3 and Iceberg tables, goes live six months after kickoff, and cuts pipeline… | ||
| Build without limits with AWS Cloud infrastructure | Tomas Mihalyi | 32m |
AWS claims hardware isolation via AWS Nitro with API‑only access, encrypted regional and backbone traffic including post‑quantum layers, and Nitro SSDs that… | ||
| Rio Tinto unlocks potential of generative AI | Benn Cizauskas, Ke Shi | 27m |
Rio Tinto builds a hybrid RAG pipeline on Amazon Bedrock using Titan embeddings, Neo4j-style knowledge graphs, OpenSearch-based vector indices, Lambdas for… | ||
| Innovations in AWS analytics: Data processing | Chris Horder, Allison Quinn | 28m |
EMR shows ~3.9x speed for Spark and ~2.7x for Trino out of the box; moving to Graviton 4 (transcript cites Graviton 3 → ~20% faster at 15% lower cost) yields… | ||
| Building with Amazon SageMaker Unified Studio | Kaushik Chatterjee | 29m |
Amazon SageMaker Unified Studio surfaces a lakehouse with Iceberg APIs, a business catalog, and a single IDE where notebooks, a query editor (Query Books) and… | ||
| Agentic AI systems using Model Context Protocol | Mithil Shah | 28m |
Demonstrates Model Context Protocol (MCP) with Amazon Bedrock agents by wiring a Bedrock inline agent to MCP servers (HTTP transport, OAuth 2.1 auth) and… | ||
| Building Netflix - from movie magic to ML reality | Nagib Shah, Luke Ballantine, Matt Byrne | 29m |
Netflix’s “It’s What’s Inside – The Game” runs real‑time face modelling (LivePortrait) and retrieval‑based voice conversion, puppeting friends’ video/audio at… | ||
| Building Agentic AI applications on AWS | Aman Sharma, Bikram Bora | 28m |
Three AWS paths for agentic apps: out‑of‑box Amazon Q, fully managed Amazon Bedrock agents, or DIY with LangChain/SageMaker. Demo builds a trip planner with a… | ||
| Building generative AI-powered products | Frank Tan, Aaron Wallis | 30m |
Lexer builds an embedded generative-AI feature called Lexi assistant that analyzes billions of customer events, using pre-aggregation and an early RAG pattern… | ||
| Master enterprise-scale generative AI | Aarthi Natarajan, Milan Pareek | 33m |
ReconAI runs in production classifying rail faults with a RAG + agentic workflow (S3, Glue, Athena, EC2, Bedrock/Anthropic Claude) and achieves 96%… | ||
| Improving responses in generative AI applications | Chiedu Eluehike, Sam Mokhtari | 28m |
Demonstrates a concrete RAG pipeline using Amazon Bedrock to auto-discover data sources, ingest documents, run embedding models into a vector store (default… | ||
| Streamlining data governance | Derrick Choo, Ram Vittal | 30m |
Concrete recipes for governing GenAI using AWS: walks through a prompt catalog plus versioning (MLflow v2.21 or Bedrock prompt manager) and recommends… | ||
| Generative AI-powered app development | Derek Bingham | 28m |
A live demo uses Amazon Q and the Amazon Q developer CLI to add a missing delete route to a simple Flask app, producing code, tests and a working local feature… | ||
| Software in the age of human and AI collaboration | Sri Nadendla | 30m |
Generative AI tools like Q developer write code, generate tests, review legacy code and run operational investigations during incidents; applied at Amazon… | ||
| Driving growth with innovative digital products | Gabriele Lasauskaite, Rad Macakanja, Tom Glover | 01h 30m |
Companies describe concrete moves: OCR with Amazon Textract and SageMaker turns 26 million pages at $1.75 per page into an automated product (Smart Do 360)… | ||
| Build an enterprise Agentic AI platform at Powercor | Peter McTaggart, Roshan Thomas | 29m |
Powercor deploys an extensible agentic AI platform on Amazon Bedrock using Claude Sonnet 3.5, a vector DB and hybrid RAG with reranking and critics to cut… | ||
| Agentic AI architecture patterns | Aaron Sempf, Andrew Hooker | 31m |
Argues that agents replace symbolic rule cortices with LLMs and that an agent is perceive‑reason‑act plus short‑ and long‑term memory, then maps concrete… | ||
| Generative AI: Path to Value in practice | Slavik Dimitrovich | 29m |
Generative AI projects show a surprisingly high failure rate at every stage because teams lack shared success criteria, pick the wrong models or DIY tooling… | ||
| The Al game: How do companies win | Chiara Longhi | 34m |
Organisations face cultural, not just technical, barriers: Harvard reports 91.9% of executives cite culture as the biggest obstacle, and AWS data shows 54% of… | ||
| Generative AI in action: from POC to business value | Shivonne Londt | 30m |
AWS shows concrete deployments: Amazon Bedrock and SageMaker run foundation models and agents on Tranium and Inferentia chips; Nova models power image, video… | ||
| Architect AI success in a regulated world | Matt Coles, Rishi Singla | 29m |
Covers a three-step approach—regulatory alignment, security controls, and compliance monitoring—with concrete patterns: classify use cases from scope one to… | ||
| Enhancing customer experience at a lower cost | Elias Sayigh, Matt Condie | 27m |
Flybys migrates its entire service stack (voice, chat, email, social) into Amazon Connect in seven months, cutting platform costs by 50% platform cost… | ||
| Simplifying data flows | Josh Tow | 29m |
Claims that Amazon tools remove pipeline plumbing: uses Zero ETL ingestion into an open SageMaker Lakehouse, ingesting relational (Aurora → Redshift), DynamoDB… | ||
| CommBank's journey with AWS data marketplace | Terri Sutherland, Olatunde Baruwa, Praveen Kumar | 29m |
Commonwealth Bank migrates 61,000 pipelines and 10 petabytes of on‑prem data to AWS in nine months with 229,000 tests, then builds a cloud‑native internal data… | ||
| How Amazon built Nova FMs using Amazon SageMaker | Pauline Kelly, Shivonne Londt | 31m |
Amazon builds the Nova family of foundation models on Amazon SageMaker HyperPod using thousands to tens of thousands of accelerators (NVIDIA H100 P4D/P5 and… | ||
| Revolutionising epilepsy care with AI/ML | Juan Mejia, Jonas Haderlein, PhD | 30m |
The Australian Epilepsy Project runs a production pipeline on AWS that cuts pre‑surgical assessment time by 70% reduction in neurologist assessment time by… | ||
| The tools and techniques behind resilient systems | Andrew Goifeld, Islam Ghanim | 26m |
A customer fleet outage hits 2,000 instances after a deployment and a startup script overwhelms an internal API, creating a retry storm; engineers show a… | ||
| Architect for ISM compliance on AWS | Farhan Farooq, Chamandeep Singh | 30m |
Outlines a practical ISM-to-AWS mapping: notes there are 900 controls and shows how 164 IRAP‑assessed AWS services map to ISM principles. Demonstrates specific… | ||
| Data classification with the Australian Government | Cain Hopwood, Stirling Jones, Aiden Kent | 30m |
Australian Bureau of Statistics builds an AWS MLOps pipeline with SageMaker pipelines and Jupyter notebooks, migrating from on‑prem 600k/week training to cloud… | ||
| Democratise data access with secure environment | Aun Iftikhar, Harmeet Chhabra | 31m |
New South Wales Health centralizes clinical and corporate systems into an enterprise data lake that grows to 500 TB and ingests about 700 GB/day, powering 100+… | ||
| Service NSW's AWS Cloud transformation story | Pranav Gothadiya, Kamaljit Bhardwaj | 26m |
Service NSW moves 200+ AWS accounts to a standardized platform and cuts spend by 40% in 18 months using cost visibility dashboards (QuickSight), right‑sizing… | ||
| DCS Spatial Services enhances emergency response | Caragh Lyons, Katie Peters | 23m |
Department of Customer Service Spatial Services moves aerial imagery processing from days to hours by rearchitecting pipelines on AWS, processing 700 km² in 5… | ||
| Future Forward Forum | Greg Willis, Colette Grgic | 01h 09m |
90 minutes of roundtable sessions use headphones and table hosts to run peer-led discussions about agentic AI across healthcare, financial services, SaaS and… | ||
| Mastering AWS Cost Optimisation | Siddh Jelapure, Cameron Yorke | 27m |
AWS announces practical cost tools and concrete rules: Compute Optimizer idle recommendations flags idle EBS (example: <1 IOPS/day over 14 days) plus… | ||
| Anthropic's Claude in Amazon Bedrock | Ben Hillman | 29m |
Anthropic’s Claude runs on Amazon Bedrock using a new MCP protocol that turns remote APIs and databases into callable tools so LLMs can fetch real context… | ||
| Data foundation for analytics and AI | Francis McGregor-Macdonald | 29m |
Cloud data stack that ties Amazon S3, S3 tables, and Apache Iceberg into SageMaker Unified Studio, claiming zero ETL for up to 15 sources and an S3-native… | ||
| Lessons from migration and modernisation journey | Evgeniy Vaganov | 29m |
Summary of concrete outcomes and tools from hundreds of migrations: a three-step method — assess, mobilize, migrate and modernize — with foundations like… | ||
| Building a sustainable future with Amazon and AWS | Bella Zanesco, Colette Grgic, Chris McInnes | 25m |
Amazon reports achieving 100% renewables for data center energy seven years early, $467M invested in Australia, and up to 94% customer carbon reduction and… | ||
| Building Generative AI features | David MacDonald, Nicola Smyth | 27m |
Startups describe concrete wins: Vital uses Amazon Nova Micro and Anthropic Claude Sonnet 3.5 to categorize ER notes with 95% accuracy, finding hundreds of… | ||
| Anti-ransomware strategies on the cloud | Loi Liang Yang, Sam Zhang | 30m |
Seven ransomware phases map to concrete AWS controls: reconnaissance blocked with CloudFront + WAF + Shield Advanced and rate/bad-IP/bot rules, initial access… | ||
| Boost employee productivity with Amazon Q Business | Brad Ryan, Ramesh Heidary | 30m |
Amazon Q Business connects to 40+ apps and your data, returns source‑cited answers, and enforces permissioned access and guardrails while never using customer… | ||
| Rethink business value for generative AI use cases | Olivier Debaillon | 32m |
Claims and numbers: $22M savings for NIB with 60% chat deflection (voice ~15%), Orion Health folds 300,000 documents into an intelligent search for ~20%… | ||
| Building advanced workflows with AWS Step Functions | Navya Srivastava, Nader Eskandari | 25m |
Shows how AWS Step Functions replace glue code with concrete patterns: use states.uuid intrinsic instead of a Lambda to save transitions, nest standard +… | ||
| From firefighting to foresight: Powering innovation | Yadi Narayana | 24m |
Datadog demonstrates how its platform moves teams from reactive firefighting to proactive operations by using Watchdog finds root cause AI to correlate logs… | ||
| Modernise using Event-Driven Architectures | Nishant Dhiman, Tomas Mihaly | 24m |
Prototype adds pre‑appointment guidance without touching the monolith by using the leave‑and‑layer pattern and AWS event plumbing: front end → EventBridge →… | ||
| Securing AI by design | Riccardo Galbiati, Marc Gollop | 28m |
Cloud security teams learn a concrete framework: discover models and all AWS connections, assess model and application posture (scan models themselves, check… | ||
| SaaS migration and modernisation | Zp Bappi, Tim Litton | 28m |
Three migration paths with concrete trade-offs: VPC-per-tenant full-stack silo (lift and ship, control plane adds onboarding and billing), layered migration… | ||
| The journey from legacy to cloud - native success | Chafic Abdallah, Claudia Howlett, Isha Dalal | 27m |
Legacy PIM running on three EC2 boxes that cost north of $250k/year gets replaced with a serverless pipeline using S3 + Transfer Family + Glue + DataBrew +… | ||
| Building database queries with Amazon Bedrock | Rita Ladda, Varun Saxena, Aman Lal | 28m |
Shows a working Text-to-SQL system that uses Amazon Bedrock plus vectorized schema embeddings (saved to OpenSearch/Aurora/RDS) and Glue crawlers to keep… | ||
| HBF’s AWS cloud transformation | Rachel Rui Liu, Billy Martin | 27m |
HBF migrates roughly 90% of infrastructure to AWS, replaces a 25–30 year mainframe core with a modern Windows-based core, and replatforms CRM to Salesforce and… | ||
| Architecting real-time transactional data lake | Masudur Rahaman Sayem | 26m |
Shows a working pipeline that writes CDC and streaming events directly from Kinesis/MSK or relational logs into Apache Iceberg on S3 using Amazon Kinesis Data… | ||
| High-performance data foundation for AI success | Jason Hunter | 35m |
Argues that a data platform must act as a business operating system and shows a concrete stack: Apache Iceberg cataloging a lakehouse, governed by AWS Lake… | ||
| Innovation through self-service data products | Rizwan Wangde, Serena Yau | 27m |
A financial institution describes how it replaces heavy custom cross-account glue code with SageMaker Unified Studio catalog to automate Lake Formation… | ||
| Industry generative AI in action | Rodrigo Castillo, Carrie Chiu, Jen Jackson | 29m |
Panel covers how Accenture and AWS build cloud foundations and an RAI-ready platform to run GenAI at scale, including Geni Genai in a box and funding via AMP… | ||
| Unleash the power of real-time data | Paul Gruber, Jeannine Farino, Quentin White | 29m |
Woodside and Versent build a cloud‑native edge‑to‑cloud pipeline that ingests plant alarms and time‑series tags directly from the OT domain into AWS via DMZ… | ||
| Building cloud security programs | Matt Preswick, Edwin Kwan | 28m |
Wiz is deployed for agentless cloud scanning across code, pipeline and runtime and aggregates findings into a security graph database to link source secrets… | ||
| How to survive a Lambda-induced data tsunami in AWS | Abhay Sharma, Alok Barsode, Michael Ewald | 26m |
AWS logging failures trigger a six-hour outage in healthcare apps after a single log entry is duplicated 80,000×, causing Lambda functions to re-ingest and… | ||
| How Seven delivers large-scale streaming success | Aron Marden, Pablo Arce | 29m |
Seven plus explains how it delivers massive live events using AWS, New Relic and CloudFront logs to correlate device ID, CDN chunk requests and DRM checks… | ||
| Realise the value of data with Splunk | Trav Kane | 27m |
Splunk demonstrates concrete tooling to avoid ingesting everything: use Ingest Actions, Edge Processor or the Splunk-managed Ingest Processor to filter… | ||
| Mitigating OWASP Top 10 CI/CD with AWS | May Kyaw, Vasanth Selvaraj | 28m |
CI/CD pipelines get concrete controls: centralize identities and packages (use AWS Identity Center, AWS Organizations, CodeArtifact), move secrets into AWS… | ||
| AWS security at scale | Tony Fendall, Igor Izotov | 28m |
AWS claims a hardware-to-edge security stack: Graviton 4 encrypts memory and hardware interfaces and Nitro enforces no human admin access with live firmware… | ||
| Generative AI driven sensitive data protection | Pratima Singh, Zohreh Norouzi | 26m |
Demonstrates an automated triage that takes Amazon Macie findings, calls Amazon Bedrock (Anthropic Claude 3.7) and SageMaker notebooks, and returns an urgency… | ||
| Application modernisation with Red Hat | Mandi Buswell | 28m |
Counties Energy migrates IBM Maximo to a managed Red Hat OpenShift Service on AWS, using Fujitsu for design and ops, and claims they avoid hiring specialist… | ||
| nothing matches | ||