AI / LLM Engineer
An LLM engineer who builds GenAI back-end features end to end: intent routing, RAG over vector search, structured-output validation, guardrails and token/latency logging on FastAPI with AWS Bedrock or OpenAI. Also has hands-on edge-AI experience, shipping a quantised OCR model as a native C++ inference library on ARM hardware.
- Level
- Mid
- Experience
- 2–3 years
- Based in
- Vietnam
- Languages
- Vietnamese (native), English (intermediate)
Selected work
- Retail operations (Japanese enterprise client): built back-end components for KPI explanation, free-chat routing and report generation on AWS Bedrock, with OpenSearch-grounded answers, fallbacks and scenario tests.
- HR tech (Japanese client): designed an LLM CV-parsing and job-fit scoring engine that blends vector similarity with rubric-based LLM evaluation and per-dimension explanations, exposed to ATS integrations.
- Internal operations: delivered a policy and HR knowledge chatbot with source-referenced RAG answers and guardrails against unsupported questions.
- Public transport (Japanese client): trained and quantised a Japanese OCR model for onboard bus displays and shipped it as a native C++ library on an ARM device to support cashless fares.
Domains
Retail, HR tech, Transportation, Internal knowledge management
Education
Engineer's degree, Information Technology