AI Engineering
Designed for practitioners building AI systems and applications in production this track explores novel architectures and system designs, from innovative approaches to RAG and agent pipelines to real-world case studies backed by production metrics.
AI Engineer Melbourne 2026 — AI Engineering (Day 1 Midday)
Wednesday 3rd June 26 12:30 pm
AI Engineer Melbourne 2026 — AI Engineering (Day 1 Midday) features Tanya Dixit (Google) on Evaluation Precedes Evolution: Rubrics as the Load-Bearing Infrastructure of Self-Improving Agents, Navan Tirupathi (Arivminds) on Beyond Forgetful Bots: Architectural Patterns for Persistent, Proactive Claw-Style AI Agents, Adam Hudson (Notion) on Shipping Sandboxed Workers for Notion Agents, Moss Ebeling (Optiver Asia Pacific) on Close your agentic loop, Anannya Roy Chowdhury (AWS) on How Many Agents Are Too Many? The Hidden Cost of Multi-Agent Systems, and Adesh Gairola (raxIT Labs) on Kill the God Agent.
Watch SessionAI Engineer Melbourne 2026 — AI Engineering (Day 1 Afternoon)
Wednesday 3rd June 26 3:30 pm
AI Engineer Melbourne 2026 — AI Engineering (Day 1 Afternoon) features Daniel Nadasi (Google) on Agent Observability: Monitoring and Understanding Agents at Internet Scale, Yicheng Guo (REA Group) on Our AI Hallucinated in Production: How We Fixed It With Evals, Abdul Karim on The Application Layer Is the New Research Lab, Stephen Sennett (V2 AI) on Orbital Lasers vs For Loops: Economically Matching Models to Tasks, Theodoros Galanos (Aurecon) on Your AI Can’t Engineer (Yet), and Michael Hart (Cloudflare) on Flue: The Agent Harness Framework.
Watch SessionAI Engineer Melbourne 2026 — AI Engineering (Day 2 Midday)
Thursday 4th June 26 12:00 pm
AI Engineer Melbourne 2026 — AI Engineering (Day 2 Midday) features Shivay Lamba (Qualcomm) on Deploying AI at the Edge: Model Compression and Hardware-Aware Optimization, Avni Bhatt on When a Small Language Model Beat Our LLM in Production, Jack Rudenko (MadAppGang) on Multi-Model Collaboration with Claude Code: How to Measure What Actually Works, Jeremy Kelaher (SBS) on Edge AI with Direct Device Control, Matthew Gillard (V2 AI) on COBOL and AI: Building a Self-Serve Knowledge Layer for 2,000 Batch Jobs, and Chris Rickard (Userdoc) on Legacy Software + Agentic Discovery.
Watch SessionAI Engineer Melbourne 2026 — AI Engineering (Day 2 Afternoon)
Thursday 4th June 26 3:00 pm
AI Engineer Melbourne 2026 — AI Engineering (Day 2 Afternoon) features Mal Curtis (NVIDIA) on Why LLMs Fall for Stories (And 5 Production Patterns That Actually Stop Them), Pas Apicella (Snyk APJ) on Hacking the Model: AI Red Teaming in Practice, Moin Zaman (Smartnote) on Why Most AI De-Identification Fails in Production, And How We Built One Lawyers Actually Trust, Daizen Ikehara (Auth0) on Are Your AI Agents Secure? Defending the Privileged Agent, Ananya Roy (Databricks) on Your Agents Pass Every Benchmark—Then Memory Breaks Them in Production, and Lovee Jain on AI Agents Are Distributed Systems.
Watch Session