Teng-Ruei Chen
AI infrastructure · LLM inference · Deep learning · Bioinformatics
Most of what AI costs is moving from training to inference. A model is trained once and then served millions of times, on hardware that is scarce, priced differently in different places, and not interchangeable. Deciding which request runs on which machine, at what cost, is becoming a question of infrastructure economics rather than an implementation detail. That is the problem I work on, and it is the same problem I worked on in power distribution, in vehicle charging, and in compute scheduling before that.
The harder half is that these decisions are difficult to evaluate. Reported gains in routing are measured against baselines that are themselves unstable, and the hardware underneath does not reliably return the same answer twice. Until the measurement is sound, the field cannot separate a real improvement from an artifact of how it was measured, and will keep paying for the difference. That is why much of my work is measurement rather than method.
My bachelor's and master's degrees are in computer science, and my doctorate is in bioinformatics and systems biology from National Yang Ming Chiao Tung University, where I worked on protein secondary structure prediction: faster prediction without loss of accuracy, a secondary-structure-based scoring matrix, the theoretical limits of the problem, and CirPred, a structure modeling and linker design system for circularly permuted proteins. I have worked in industry throughout, and I am an inventor on twelve granted patents in compute scheduling and AI infrastructure.
- 2021
- Ph.D. in bioinformatics and systems biology, National Yang Ming Chiao Tung University, Hsinchu
Education
- 2026–
- Chief Technology Officer, BiiLabs Co., Ltd. · Taipei
- 2026–
- Chief Executive Officer, Krixvon Company Limited · Taipei
- 2024–26
- Chief Technology Officer, Taiwan Intelligence Research and Development Co., Ltd. · Taipei
- 2022–24
- Engineer, Industrial Technology Research Institute (ITRI) · Hsinchu
- 2018–22
- Chief Technology Officer, Lingshing Intelligence Co., Ltd. · Taipei
Positions
Routing
Learned routers assign requests across a pool of models. I measure how far those assignments sit from an oracle that knows answer quality in advance.
Privacy
Serving infrastructure emits metadata before the response exists: which tier was picked, how long the queue was, how many drafts were sampled. That channel is readable.
Validity
Whether routing results measured on benchmarks survive contact with production traffic shapes, and what breaks first when they do not.
Publications
Work before 2022 is protein structure prediction. The methodological question carries over.
Datasets & software
Patents as inventor
| Region | Number | Title | Status |
|---|---|---|---|
| TW | — | Scheduling system and method combining cloud and local computing power | Under examination |
| TW | — | Battery Swapping Scheduling System and Scheduling Method | Allowed |
| US | — | Global cross-timezone computing power scheduling method | Allowed |
| TW | I941269 | Regional computing power management platform system | Granted |
| TW | I935869 | Global cross-timezone computing power scheduling method | Granted |
| TW | I928765 | Computing power demand analysis system and method thereof | Granted |
| TW | I928714 | An open analytical structure combining depth and breadth with dual engines of thoughts | Granted |
| TW | I907964 | Power information processing system and method based on blockchain | Granted |
| TW | I882792 | Intelligent vehicle-assisted drone operation system | Granted |
| TW | I870046 | Wireless charging system and method for unmanned vehicles | Granted |
| TW | I829017 | The system and the method of battery exchange of the electric car | Granted |
| TW | I827983 | The system and the method of wireless charging of the electric car | Granted |
| TW | I828105 | Logistics processing system and method thereof | Granted |
| TW | I824753 | Charging service system and method thereof | Granted |
| TW | I741788 | A wide area power monitoring platform system and method | Granted |
Listed as inventor; rights are held by the respective assignees. Titles as recorded by the patent office.
Awards
- Honorable MentionIntelligent production scheduling systemAIGO
- ExcellenceAI hair-region detectionAIGO
- Honorable MentionAI planning and scheduling systemAIGO
- Judges' AwardUAV-based human detection systemDigital Technology Solution
- Judges' AwardAI fish-species recognition systemDigital Technology Solution
- OutstandingDynamic prediction model for hospital outpatient waiting timesAIGO
- WinnerKoi image recognition systemAI Application Challenge
- ExcellenceESG extraction and sustainability-indicator matching modelAI GOOD
- Judges' AwardGenerative AI for varied interpretation of aquarium exhibitsDigital Technology Solution