Machine learning · Computer vision · LLM systems
I build machine learning systems and ship them. PhD from the University of Washington, then IBM Research, Amazon, and since 2025 co-founder and CTO of FluenTea, where I designed and built the product, from the voice agent to the app stores.
FluenTea is a video classroom and AI speaking coach for language schools. I co-founded it in 2025 and built it myself, from the first commit to the app stores. The science is in making speaking measurable and the coach trustworthy. Every class and practice call is transcribed, and from the transcripts we compute speaking time per student, the grammar each student used and struggled with, and what the teacher should reteach next lesson. The coach, ChaCha, follows a written behaviour spec that sets how hard she pushes from the learner's level and state. Her answers are tested three ways. Prompt-side tests check what she was told before she spoke. A red-team suite plants injected text in every path. A live suite graded by a judge model keeps a per-kind hallucination baseline that may only improve. In the first university pilot, most students kept practising with her beyond what their teachers had asked for.
The engineering around that is the web app, the backend and the Kubernetes cluster, the Android and iOS apps, payments, and the consent and safety work a university required before enrolling minors. We launched three months after the concept, have more than 16,000 registered users in 136 countries, and work with three universities, the first of them in Mexico with 500 students.
Before that I was an applied scientist at Amazon in Berlin for five years, working on computer vision and multimodal models that run on the retail site, and a research staff member at IBM Research Almaden, where I worked on getting machines to read tables in documents. My PhD was about decoding what people are doing from recordings of their brains and video of their behaviour.
I co-founded FluenTea and built the whole product myself, from the voice agent to the app stores. Language schools run live classes in it, an AI coach sits in and runs practice calls between classes, and teachers are told what to reteach.
Technical lead on three computer vision projects that went to production on amazon.com and Alexa. One reads product dimensions out of images with a text-VQA model and filled in sizes for more than ten million products; another predicts the interests in creators’ photos and videos with CLIP and feeds two recommenders. Door Desk award; D-Extract at WACV 2023.
At IBM I led GTE, a table extraction system (WACV 2021), built the TableLab labelling tool around it, and organised a SemEval 2021 shared task on table fact checking. My PhD work on decoding behaviour from brain recordings and video was an oral at AAAI 2018.
I love brains and computers, basically any cognitive system. I was born in Nanjing, moved to Canada at nine, studied in Vancouver and Seattle, worked in San Jose and Tokyo, and have lived in Berlin since 2021, where I am slowly learning German. I like living inside other cultures, which is probably also why I like role-playing and strategy games (Dragon Age and XCOM most of all).