I do not know Jason yet.
Start with the Now page, then choose Research, Systems, or Writing.
研究與寫作
這是 Jason Chia-Sheng Lin 的個人研究網站與寫作空間:整理近期在可信任 AI、語音決策穩定性、臨床流程支援、AI agent governance、資安與 AI 系統工程教學上的公開安全成果。
Recent work is summarized from public-safe repo artifacts and planning summaries. Raw planning notes, private contacts, patient-like data, raw transcripts, credentials, and patent-sensitive mechanics are intentionally kept out of the website.
Start with the Now page, then choose Research, Systems, or Writing.
Start from research programs, public artifacts, and validation paths.
Open the Systems page and inspect each workflow by capability, evidence, and next validation layer.
Use Teaching and Writing for long-form notes, course surfaces, and public framing.
Packaged the AI Systems Engineering Handbook into a 7-day consulting-style onboarding path for enterprise voice AI, AI Gateway, agent governance, red teaming, K8s, GPU sizing, and customer acceptance evidence.
Advanced a speech-to-decision stability research line that tests whether downstream decisions remain stable under plausible ASR alternatives.
Developed provenance-bounded evidence graph packet work for small-model decision support, with release gates, audit protocols, and public-benchmark orientation.
Built a synthetic vital-aware intake and staff-review summary demo for a June market demonstration, with a narrow API contract and explicit clinical-scope controls.
Hardened a urology previsit workflow prototype for adaptive question navigation, missing-field repair, clinician-review summaries, and PSA/CRM-ready proposal framing.
A speech-to-decision stability research line that asks whether downstream decisions remain stable under plausible transcript alternatives.
Research on evidence graph packets that help small models make more stable, better-grounded decisions under the same token budget.
A technology-governance research line on how action-capable AI can appear governable while reconstructability collapses underneath.
A security-evidence research direction that turns reviewer feedback into traceable public-source case evidence, uncertainty labels, and baseline comparison.