Cake Talent Search

in current technologies, trends and programming languages. Key points are: 1. For last 2+ year working on FinTech Product worked multiple APIs, Kafka, User Retention based on ML/AI 2. Prior full-time working company was Infobeans Indore which worked on ML Document classification using NLP. 3. Worked in HighIq, Hyderabad as ML expert doing projects on Tesseract, Multithreading, OCR, Doc Classification, Windows Services-Polling, NLP. 4. At Emorphis as Python Developer, worked on IoT project. I also worked on an Audio project using Tensorflow, Keras, and scikit. Done

抓取金融新聞，將繁體文章標題與內文透過語言轉換工具(opencc)轉為簡體，最終使用情感分析工具(snownlp)進行股市漲跌分析。 技術：NLP、Beautifulsoup、 opencc 、 snownlp 、 Matplotlib 台灣景氣指標預測模型 side project 使用深度學習框架(Pytorch)自行搭建預測模型，以各式台灣景氣指標當作輸入，輸出未

and improving development efficiency. Optimized website performance with server-side rendering, caching, and resolving critical issues like Algolia-related server-side crash. Mentored team members on React best practices, fostering a collaborative and skilled team environment. Research Assistant, Academia Sinica 11//2020 Conducted research in NLP and machine learning, focusing on models like BERT and multi-task learning. Authored a publication: "Detecting Deceptive Language in Crime Interrogation," presented via a recorded session at an NLP conference. Developed a Memory Networks-based application for detecting deceptive language in collaboration with an international research team

speed by 20%. - Designed a high-performance data warehouse on AWS, improving cross-team data accessibility and reducing query times by 35%. EDUCATION National Taipei University of Technology • Master’s Degree in Electronic Engineering MaySep 2019 | Taipei, Taiwan • Specialized in developing deep learning models for Natural Language Processing (NLP). • Developed classification systems for speech/audio using deep learning techniques. • Conducted research on image classification and recognition through deep learning methodologies. CERTIFICATIONS • Google IT Automation with Python, Coursera - FebPATENT • ABNORMAL SOUND DETECTION METHOD AND APPARATUS (USA1) - FebSKILLS Programming & Frameworks : Python, TensorFlow, Keras, PyTorch, scikit-learn

參與各種專業講座、研討會，深入研究了最新的人工智慧技術和趨勢，擴展了我的專業知識，特別是在機器學習、深度學習和自然語言處理等領域。從中獲得的見解對我未來的工作和研究產生了積極影響 🏆 畢業生代表 2019 北科服務表現 績優獎 熱

行人工標註和確認。 知識圖譜：維護現有的關聯資料，開發API使其可以查詢、修改、合併。 智能營銷工具：(1)將經過命名實體識別、自然語言處理、分詞和情感分析等前處理後的文章進行算法提取可能的聚類信息；另外，將文章提到的圖片和影片下載存

toolkits such like KALDI, ESPNET, familiar with both hybrid and end-to-end ASR architectures Speech Signal Preprocessing, Decoding, Rescoring, Reranking Large Language Model Fine-tuned LLAMA2, LLAMA3, phi-3, gemma models using FSDP-QLoRA technology Retrieval-Augmented Generation (RAG) Familiar with Transformer-based Language Models BERT, ALBERT, RoBERTa GPT-3, InstructGPT NLP Tasks Language Modeling Part-of-Speech Tagging, Named Entity Recognition Projects A BERT-based Language Modeling Framework https://www.isca-archive.org/interspeech_2022/chien22b_interspeech.html Proposes a Mixture-of-Experts language model architecture that automatically...