A curated account of Larry’s documented work. This guide will continue its walk when you close the conversation.
Read résuméThe documented case study and general technology explanation are labelled separately.
ENGINEERING CAPABILITIES
Technical stack.
Tools and practices used across enterprise AI, cloud engineering and data platforms. Individual experience panels show where and how they were applied.
AI & agent systems
Solution design, orchestration, fine-tuning and evaluation.
LangGraphLangChainCrewAIAWS BedrockAWS AgentCoreMCPPydanticHugging FaceLoRA / QLoRALLM-as-a-Judge
Cloud & architecture
Secure deployments, model operations and infrastructure.
AWSAzureSageMakerDockerECREKSMLflowTerraformCloudFormationStep Functions
Data engineering & retrieval
Reliable pipelines and evidence-grounded information retrieval.
OpenSearchPostgreSQLHybrid search / MMRRerankingTable-aware chunkingDatabricksAzure Data FactoryAWS GluePySparkAirflow
Engineering, security & delivery
APIs, authentication, observability and cross-functional execution.
PythonFastAPIWebSocketOAuth2 / PKCEPingID SSOJenkinsLangSmithLangFuseAgile / ScrumClaude Code / Codex
This is a career-wide overview, not a claim that every technology was used at every employer.
MEET LARRY
Weixuan
(Larry) Jiang.
AI Technical Architect & Senior AI Engineer
I design and implement enterprise AI agents, secure integrations and cloud platforms. My work spans banking, healthcare and workplace safety, from architecture through production delivery.
- Experience
- 5 years across AI, cloud and data
- Based in
- Melbourne, Australia
- Work rights
- Australian Citizen
- Independent project
- Technical Founder, CoinBell AI
Original architectural interpretations. Professional experience and outcomes are grounded in my CV.
Experience
2021 — 2026Building across enterprise teams, products and platforms.
Aug 2025 – Sep 2026
Commonwealth Bank
Senior AI Engineer
Contractor via Rackspace TechnologyMulti-agent banking systems, policy-grounded RAG and production-grade AI operations.
More about this role
- SSO authentication sidecar using OAuth2, PKCE and PingID for MCP servers
- AWS AgentCore deployment, tool and model call limits, retries with backoff and a three-model fallback chain
- Release-readiness agents supporting two monthly releases across 10+ service domains and four federated business units
- Evaluation frameworks covering retrieval precision and recall, hallucination detection, traces and human feedback
Feb 2025 – Aug 2025
Amazon Web Services
Cloud Engineer — AI/ML
Melbourne · Full timeEnterprise AI/ML architecture, generative AI proofs of concept and production support.
More about this role
- Resolved complex SageMaker and Bedrock production incidents
- Invoice-processing PoC with Bedrock and Textract reduced manual effort by 80% and operational costs by 87%
- Built a restaurant recommendation prototype with dynamic menu ranking and personalisation
Sep 2024 – May 2026
JR Academy
Generative AI Instructor
Contract · Weekends & after-hoursHands-on training in LLM fine-tuning, RAG and agent-based systems for engineers and data scientists.
More about this role
- Advanced training in LoRA, QLoRA, quantisation and distributed fine-tuning
- Hugging Face PEFT, PyTorch FSDP and vLLM
- Mentored professionals from data ingestion through deployment
Apr 2024 – Oct 2024
Forwood Safety
AI Technical Architect
Australia · Remote · Full timeEnd-to-end GenAI architecture for workplace risk analysis and decision-making.
More about this role
- Expert-validated risk insights with 80–90% accuracy
- S3, embeddings and OpenSearch pipelines improved retrieval accuracy by 30% and cut processing time by 40%
- SageMaker, ECR, EKS and MLflow pipelines for reproducible delivery
- Led cross-functional teams and aligned technical solutions with senior stakeholders
Aug 2022 – Apr 2024
Accenture Australia
Senior Analyst, Data Engineer
Melbourne · Full timeHealthcare GenAI, medical coding models and reliable cloud data engineering.
More about this role
- MLOps pipeline for fine-tuning and deploying a Llama 2 7B medical coding model for ICD-10-AM
- RAG and hybrid search improved relevance, precision and recall by 30%
- Databricks and Azure Data Factory ingestion, plus AWS Glue and PySpark ETL with 99.5% data accuracy
- Step Functions workflows reduced manual intervention by 95%
Jul 2021 – Aug 2022
National Australia Bank
Analyst, Data Engineer
Jan – Aug 2022 · Previously Intern Cloud EngineerStreaming ingestion, automated ETL and cloud infrastructure supporting banking platforms.
More about these roles
- Data Engineer, Jan–Aug 2022: maintained 99% pipeline uptime with PySpark, S3, Redshift, Python and Databricks
- Automated ETL across billions of rows, saving 20 hours of manual work per month
- Khoros API pipelines saved 12 hours of manual work weekly
- Cloud Engineering Intern, Jul 2021–Jan 2022: Terraform, Jenkins build triggers, EC2 backups and database refresh workflows
Selected work
01 / 04COMMONWEALTH BANK · 2025–2026
Covenant intelligence
A policy-grounded multi-agent system that turns business loan assessment from weeks into hours.
Weeks to hoursBusiness loan assessment cycles
Ask about this projectView contribution & evidenceMy contribution
Designed a four-node agent workflow with parallel Covenant and Undertaking agents and shared industry context.
LangGraphOpenSearchPydantic
Visit CoinBell AIArchitecture & technical details
- Four-node LangGraph DAG with parallel Covenant and Undertaking agents and a shared Industry Analyst
- Two-tier OpenSearch RAG with table-aware chunking, reranking and policy-cited, structured outputs
- Human feedback and LLM-as-a-Judge across 20+ industry verticals
Policy knowledge
Covenant agentUndertaking agent
Cited recommendations
Shared industry context · Structured outputsExpertise
01AI architecture
Multi-agent orchestration
RAG & hybrid search
LLM evaluation & fine-tuning
Stakeholder & solution design
LangGraph · Bedrock · AgentCore
LangChain · CrewAI · MCP 02Production systems
Secure service integrations
Reliability & model fallback
MLOps & observability
Containerised deployment
SageMaker · EKS · MLflow
OAuth2 · FastAPI · Terraform 03Data platforms
Cloud data architecture
ETL & ingestion pipelines
Semantic & table-aware retrieval
Streaming & workflow automation
OpenSearch · PostgreSQL · PySpark
Databricks · Airflow · AWS Glue EDUCATION
Bachelor of Data Analytics
University of South Australia
Jul 2019 – Oct 2023SELECTED CERTIFICATIONS
AWS · Machine Learning, Data Engineering & AI
Google Cloud · Architecture, ML, Data & Development
Microsoft Azure · Data Engineering
Databricks · Apache Spark
View all certifications
- AWS Certified Machine Learning Specialty, 2025
- AWS Certified Machine Learning Associate, 2025
- AWS Certified Data Engineering Associate, 2025
- AWS Certified AI Practitioner, 2025
- Google Cloud Certified Cloud Architect, 2023
- Google Cloud Certified Professional Cloud Developer, 2023
- Google Cloud Certified Professional Database Engineer, 2023
- Kubernetes and Cloud Native Associate, 2023
- Google Cloud Certified Professional Machine Learning Engineer, 2022
- Google Cloud Certified Professional Data Engineer, 2022
- Microsoft Certified: Azure Data Engineer Associate, 2022
- Databricks Certified Associate Developer for Apache Spark 3.0, 2022