Po-Lin (Berlin) Chen
curriculum vitae
Education
| Aug 2025 - Dec 2026 | Master of Science in Computer Vision (MSCV) @ Carnegie Mellon University (CMU) |
|---|---|
| Robotics Institute, School of Computer Science | |
| - Coursework: Advanced Computer Vision, Deep Learning Systems, Robot Learning | |
| Sep 2020 - Jun 2025 | Bachelor of Science in Biomechatronics Engineering @ National Taiwan University (NTU) |
| Artificial Intelligence Specialization Program | |
| - GPA: 4.04/4.3, CS-Related: 4.22/4.3, Rank: 1/66 | |
| - Coursework: Deep Learning for Computer Vision, Digital Image Processing, Microcontrollers, Automatic Control | |
Work Experience
| Jun 2024 - Sep 2024 | AI Software Engineer Intern @ Advantech Edge AI Inference Team |
|---|---|
| - Optimized heterogeneous AI inference on Intel CPU/iGPU by leveraging OpenVINO/VNNI acceleration; achieved 2.7× faster inference and 2.2× E2E speedup, establishing a new performance standard for a novel CNN architecture | |
| - Designed high-reliability anomaly detection by augmenting vision models via multi-modal input channels; demonstrated mission-critical system reliability (99% precision, 2% false alarm rate) across 200+ production trials | |
| - Standardized edge deployment pipeline by containerizing and delivering 10+ supporting modules (APIs/debug/telemetry) into the C++ inference engine, enabling high-throughput API services and management for decentralized 6-8 device clusters | |
| Sep 2022 - May 2024 | Applied ML Research Engineer @ Biophotonics & Bioimaging Lab |
| - Architected Edge-Cloud ML System using Flask and AWS (IoT Core, S3, Athena); built CI/CD pipeline with GitHub Actions for over-the-air (OTA) software updates and IoT Core/MQTT for automated performance profiling | |
| - Drove resource-constrained deployment on Raspberry Pi with LiteRT/TFLite; quantified performance tradeoffs across inference optimization strategies, achieving 63% latency reduction and 40% RAM cut for real-time, low-power operation | |
| - Delivered QuickSight reporting framework by collaborating with domain experts, saving 84% operational labor overhead | |
| Aug 2024 - Dec 2024 | Teaching Assistant @ Dept. of Biomechatronics Engineering, NTU |
| - Introduction to Artificial Intelligence (Prof. Chien-Yu Chen) | |
| - Computer Programming Language (Prof. Ta-Te Lin) | |
| - Practical Data Structures and Algorithms (Prof. Chien-Yu Chen) |
Projects
| 2025 | GPU-Accelerated 3D Perception Framework | HPC, 3D Vision, Robotics Systems |
|---|---|
| - Re-implemented ORB extractor and image pyramids with custom CUDA kernels, achieving 2.8× speedup in tracking | |
| - Performed Nsight GPU profiling on the custom implementation against OpenCV on autonomous driving datasets | |
| 2024 | Controllable & Personalized Text-to-Image Diffusion | Vision-Language Models, Generative AI |
| - Accelerated style-conditioned multimodal diffusion training by 44% via Image Prompt Adapters and ControlNet | |
| - Enhanced PEFT/LoRA model fidelity by introducing structural separation loss to resolve attribute leakage; outperformed SOTA LoRA-based diffusion by 1.4% on CLIP-I and 2.2% on CLIP-T in multi-concept text-to-image personalization | |
| 2023 | Greenify: Your Recycle Mentor @ Google Hardware Product Sprint '23 | Embedded ML, Generative AI |
| - Attained 95% accuracy in autonomous waste identification by few-shot learning with domain-specific image data | |
| - Demonstrated <150ms low inference latency with a computation-efficient LiteRT-based Flask backend on Raspberry Pi | |
| - Boosted user engagement with 25+ AI-generated animations, 20+ immersive storylines, and an interactive touch interface | |
| - Earned Product for Social Good & Best Product of the Year from Google Taiwan | |
| 2021 - 2022 | Robot Perception & Navigation @ Field Robot Competition '21 & '22 | Multimodality, Embedded ML, Robotics Systems |
| - Performed multimodal sensor fusion (vSLAM + ultrasonic) to enhance system resilience for near-range motion planning | |
| - Optimized YOLO inference on Jetson TX2 using TensorRT, achieving 4.7× throughput speedup and 92% success rate |
Awards
| 2024 | Short-Term Research and Advanced Studies Grant |
|---|---|
| - $1,000 from Lin Penyuan Chinese Culture Education Foundation for attending ISMAB at Bali, Indonesia | |
| 2024 | Best Oral Presentation Award |
| - Machine Learning Topic Group at 2024 CIGR International Conference, Jeju, Korea | |
| 2023 | Undergraduate Research Fellowship |
| - $1,600 from National Science and Technology Council | |
| 2023 | Best Product of the Year & Product for Social Good |
| - Google Taiwan | |
| 2022 | Dean's List Awards |
| - National Taiwan University, 2021 Fall & 2022 Spring | |
| 2021 | Top 20% Robot Development Team |
| - Field Robot Competition, National Pingtung University of Science and Technology |
Extracurricular Activities
| 2023 | International Student Volunteer @ NTU Office of International Affairs (OIA) |
|---|---|
| - Organized campus tours and orientation sessions for newly enrolled international student | |
| - Provided one-on-one assistance to Indonesian and Colombian students, addressing queries on registration, funding, academics, and campus life | |
| 2022 | Workshop Instructor @ Field Robot Training Program |
| - Delivered lectures on best practices in collaborative system development and robotics project management | |
| 2022 | Director of Department Expo @ NTU Azalea Festival |
| - Led a team of 60 to organize the exposition, promoting prospective students’ understanding of the department | |
| - Adapted to COVID-19 constraints by leading the shift to an immersive VR showroom, attracting 15,000+ site visits | |
| 2022 | Program Director @ NTU Ukulele Winter Camp |
| - Launched an 1-week program with 26 volunteers to deliver free ukulele sessions for 50+ underprivileged students | |
| - Facilitated a donation of 50 ukuleles to under-resourced schools in rural Taiwan |
Publications
| Rui-Yuan Liao, Po-Lin Chen, Ta-Te Lin. ”Calf Health Monitoring System Based on Internet of Things and Computer Vision”. In 2024 Conference on Agricultural Machinery and Bio-Mechatronics Engineering (AMEBE 2024), Taichung, Taiwan. [Paper] |
| Po-Lin Chen, Rui-Yuan Liao, Jih-Tay Hsu, Ta-Te Lin. "Efficient AIoT Framework for Vision-Based Behavior Monitoring in Livestock". In 11th International Symposium on Machinery and Mechatronics for Agriculture and Biosystems Engineering (ISMAB 2024), Bali, Indonesia. [Paper] [Presentation] |
| Po-Lin Chen, Jih-Tay Hsu, Ta-Te Lin. "Video-Based Behavioral Assessment in Dairy Calves Based on Spatio-Temporal Convolutional Neural Networks". In 6th CIGR International Conference (CIGR 2024), Jeju, Korea. Best Oral Presentation Award [Abstract] [Presentation] |
| Po-Lin Chen, Ta-Te Lin. ”Dairy Calf Action Recognition and Analysis Based on Spatio-Temporal Convolutional Neural Network”. In 2023 Conference on Biomechatronics and Agricultural Machinery Engineering (BEAME 2023), Yilan, Taiwan. [Poster] |
Skills
| Programming Languages | Python, C/C++, Java, C#, SQL, TypeScript, Shell, CUDA |
|---|---|
| AI/Machine Learning | PyTorch, TensorFlow, JAX, Numpy, OpenCV, Sklearn, LiteRT, ONNX, OpenVINO, TensorRT; Computer Vision, Reinforcement Learning, LLMs, VLMs, Multimodal ML, Generative Models, Edge ML Optimization |
| Software/Tools | Docker, Linux, Git, ROS2, Flask, React, REST APIs, CI/CD (GitHub Actions), Azure, AWS |