AI Innovator Award 2026
Jonathan Owen
Owen builds production AI systems, OpenClaw-powered enterprise automations, and scalable full-stack platforms for teams that need reliable intelligence in the real world.
He works at Sea Limited (NYSE: SE), whose Shopee is the largest pan-regional e-commerce platform in Southeast Asia and Taiwan.
Sea Limited (NYSE: SE)
Production systems at Southeast Asia scale
DAPER / APE creator
Adaptive AI framework and production execution model
OpenClaw engineer
Enterprise automations and agent communication pipelines
6+ years
AI, cloud, full-stack, and automation delivery
Capabilities
AI engineering that reaches production.
Owen works across the product, model, workflow, and infrastructure layers so automation can be shipped, measured, debugged, and improved.
Agentic AI systems
LLM agents, RAG pipelines, tool-using workflows, and production controls built for traceable outcomes.
OpenClaw automation
Enterprise automations, agent-to-agent communication, and self-evolving agents that learn, adapt, and act proactively.
Full-stack products
Fast, maintainable interfaces and APIs across Next.js, React, TypeScript, Python, Java, and distributed services.
Cloud infrastructure
AWS, GCP, CI/CD, containers, orchestration, observability, and deployment patterns for resilient delivery.
Frameworks
DAPER for adaptive AI. APE for production execution.
Owen designed DAPER, a revolutionary Detect-Analyze-Plan-Execute-Reflect framework for AI systems that need to observe, reason, act, and improve. APE is the production-optimized variant that keeps complex agent work debuggable through clear Analyze, Plan, and Execute checkpoints.
Detect
Identify signals, risks, inputs, and state changes before work begins.
Analyze
Extract relevant context, reduce noise, and produce structured evidence.
Plan
Prepare focused knowledge, retrieve context, and define the next action.
Execute
Make the decision, complete the task, and expose reasoning for review.
Reflect
Learn from outcomes so the system can improve future decisions.
Traceable checkpoints
Intermediate outputs make agent decisions easier to inspect, test, and improve.
Focused context
Planning filters knowledge before execution so agents avoid overloaded prompts.
Reflection loops
Systems can learn from outcomes and adapt without turning into opaque black boxes.
OpenClaw Background
Enterprise automations with agents that can coordinate, learn, and act.
Owen's OpenClaw work focuses on turning AI agents into practical operating systems for business workflows: event-driven automations, agent-to-agent communication pipelines, and proactive agents that adapt as they gather new evidence.
This background supports broader AI automation work while giving OpenClaw-heavy projects a clear technical foundation.
Enterprise workflow automation
Agent-to-agent communication
Self-evolving proactive agents
Selected impact
Anonymized outcomes from real systems.
The details stay public-safe, but the shape of the work is the same: complex inputs, high-volume data, measurable business impact, and maintainable engineering.
High-stakes classification automation
Designed AI-assisted classification systems with auditable reasoning paths and production-ready safeguards.
Semantic knowledge retrieval
Built vector search architecture that turns large historical archives into fast, relevant insight retrieval.
Multi-agent research workflow
Created specialized agent roles that divide complex research into cleaner, faster, higher-quality outputs.
Trading and operations systems
Delivered automated systems for data-heavy decision loops, backtesting, execution, and operational workflows.
Technical stack
Hands-on across AI, product, systems, and infrastructure.
The stack changes by problem. The constant is ownership across the full path from prototype to production.
AI Engineering
Systems
Product
Infrastructure
Collaborations and conversations