Heewon Ko
Summary
I am a cloud support engineer on the BigData profile at AWS, working across Amazon Bedrock, Bedrock AgentCore, and SageMaker with customers from major banks to global technology companies.
I believe you get further by working backwards from where things need to end up, and by looking at the whole forest rather than the one tree in front of you. A narrow fix usually just creates the next problem. I like seeing a system as a structure and getting the parts to work in harmony, and people are much the same. I listen before I decide, I would rather build with others than alone, and when a group needs someone to pull it together I am happy to be that person, because that is where the energy comes from.
Agentic AI excites me most, because it touches all of this. I am curious about what is coming next and always happy to learn it.
Experience
- Resolving live customer cases across Amazon Bedrock, Bedrock AgentCore, and SageMaker for regulated customers including major banks, financial services firms, and global technology companies. The work spans model access and quotas, RAG quality, latency root cause analysis, agent gateway and identity, and policy configuration.
- Reproduce before answering. Every case that is not trivial gets a sandbox reproduction, CloudTrail evidence from the customer account, and a small Python tool when the behaviour is not obvious from the documentation.
- Raise documentation gaps and route service team escalations, and correct inaccurate auto generated case drafts before they reach a customer.
- Building an agentic workflow for case work to speed up case investigation.
- Writing a SageMaker knowledge article to close a documentation gap I ran into on a case, so the next engineer does not lose the same day to it.
- Contributing a documentation update for Bedrock AgentCore.
- Participated a concept proposal reimagining AWS Support as a persistent, context aware customer workspace, grounded in customer research data, with a de identified fleet pattern design and a consent based automated fix flow.
- Technical Assistant at the AWS Sydney Summit Kiro Workshop (May 2026), giving hands on guidance to customers on a newly launched tool at a public AWS event.
- Taught AI fundamentals and gave a career talk to Year 10 students, SWS Activate program (Jun 2026).
- Member of the Asians at Amazon affinity group and part of the team organising the community's Mid Autumn Festival event.
- Designed the winning Women in STEM logo and the SE team banner for APJ Tech Summit 2025.
- Completed intensive structured training in AWS core services, networking fundamentals, security, Linux administration, and Python.
- Earned five AWS certifications in 7.5 months (Apr to Nov 2025) while carrying program work, by studying before and after hours.
Earlier Career
- Six years in design roles at Design In and Out, Sydney Forró Dance, and exhibition coordination, delivering client facing work with measurable results.
- Junior Software Developer at NoHarm, an AI for health institute, contributing Python and Flask code to the product API. Retrained through Certificate IV in Information Technology at TAFE NSW, which is what moved me from designing interfaces to building the systems behind them.
- The design years are the reason I start from the customer's experience and can explain a technical idea visually when words are not landing.
Education
Certifications
AWS Certified: Solutions Architect – Associate (Jun 2025) · Data Engineer – Associate (Nov 2025) · Machine Learning Engineer – Associate (Nov 2025) · AI Practitioner (May 2025) · Cloud Practitioner (Apr 2025)
In progress: AWS Certified Generative AI Developer – Professional
Skills
Programming: Python as my working language for boto3 diagnostic tooling and sandbox reproductions, plus object oriented fundamentals in C# and Java, JavaScript, and SQL
Foundations: Linux, networking (TCP/IP, DNS, TLS), security fundamentals
Customer: technical case correspondence, root cause analysis, explaining complex issues to both technical and non technical audiences
Languages: English, Korean (native), Portuguese (conversational)