Ying (Jenny) Jiang

Ying (Jenny) Jiang M.S. in Computer Science (MSCS) School of Computer Science Carnegie Mellon University Email: yingj2 [at] cs [dot] cmu [dot] edu GitHub | Linkedin | Curriculum Vitae Actively seeking entry-level software development / ML engineering positions starting in Dec 2022!

About me

I’m pursuing my Master of Science in Computer Science at CS department, Carnegie Mellon University with QPA 4.16/4.33 and I’ll graduate in Dec 2022. During my time at CMU, I did research with Dr. Andrew Crotty and Prof. Andy Pavlo on self-driving databases, and I interned at Hudson River Trading (HRT) this summer. I earned my bachelor's degree in CS with highest honors from Yuanpei College, Peking University. I possess a background in both systems and ML engineering.

At HRT I wrote a fully functional resource reservation service and a skeleton for a market simulator. Before my master's study with straight A's, I worked full-time at Tencent WeChat, an IM app with 1.2 billion DAU in China, for a year as a full-time backend developer. During my time there, I wrote over 50k lines of production code in C++ and developed many backend skills such as server development and optimization. I previously completed internships at Apple and Megvii (Face++) where I launched computer vision algorithms to products and gained experience in both engineering and machine learning. I formerly published a paper on computer vision at CMU.

With a mixed background in both CS and AI, I came to realize the huge gap between doing theoretical research and launching a realistic program. My experiences ignited my desire to develop practical systems as an SWE for my career, realizing my self-worth through system development and deployment. As a future engineer, developing usable solutions for corporations gives me a real sense of achievement.

I'm actively seeking entry-level SWE/MLE positions starting Dec 2022. If you're interested, please don't hesitate to contact me.

Education

• M.S., Computer Science, School of Computer Science, Carnegie Mellon University, Aug. 2021 - Dec. 2022, 4.22/4.0

  • Courses: Database Systems (A+), Distributed Systems (A), Machine Learning (A+), Special Topics in Database Systems (A), ML with Large Datasets (A+), Probability and Computing (honor track) (A)

  • Selected Projects: Twitter's three-tier architecture w/ cache consistency and sharding (Go); Distributed Raft protocol (Go); TCP built over UDP (Go); Database memory manager, hash index, query executors w/ concurrency control (C++)

  • Courses in progress: Storage Systems, Parallel Computer Arch and Programming, DL Systems

• B.S. (Hons), Computer Science and Technology, Yuanpei College (an elite undergraduate program), Peking University, Sept. 2016 - Jul. 2020, 3.80/4.0

  • Ranking: Top 5%, Top 1% in Women

  • Courses: Operating Systems, Computer Networks, Parallel and Distributed Computing

  • Teaching Assistant: Introduction to Computer Systems (from CMU 15-513) for 2 years

Experiences

• Hudson River Trading, Core Developer Intern, May. 2022 – Aug. 2022

  • Designed a generic printing feature for the BATS protocol in C++20 using templatization; reduced over 80% of code for customized printing and improved flexibility of parsing

  • Completed a functional resource reservation gRPC server using Trio (async Python), a Slack client and integrated the system into HRT’s testing framework; resolved reservation inconsistency and improved developer happiness

  • Constructed a market simulator with different behaviors which would be widely used to improve testing efficiency

• Carnegie Mellon University - School of Computer Science, Research Intern, Jan. 2022 - May. 2022

  • Collaborated with 2 to build an action enumeration framework to generate candidates for the search and scoring modules in the self-driving DB system NoisePage, which restricts the search space and provides hints for planning

  • Built a prototype to evaluate the effectiveness of different database index data structures in the Apache Arrow (a columnar memory format) codebase

• Tencent WeChat, Software Engineer, Sept. 2020 – Jul. 2021

  • Designed and reconstructed the retrieval platform in WeChat’s recommendation system as main developer (36k lines of code in C++), which is achieving 40k QPS and is under huge growth

  • Developed a push notification feature using C++ and memory queue for “Top Stories” of WeChat, consisting of offline selection, online planned execution and speed control; increased 25% of users to click in and enjoy

  • Optimized and maintained over 10 modules in the recommendation backend (60 million DAU), such as relativity scoring (6 billion queries per day); enhanced parallel efficiency by over 10% and reliability toward spikes

• Tencent Cloud, Software Engineer Intern, Jul. 2020 – Sept. 2021

  • Built a controlling platform capable of deploying, configuring and rolling back AI models with Go, databases (MySQL, Redis) and CI/CD pipelines; saved developers’ time by over 70%}

  • Collaborated with 8 to support new features of Tencent Cloud's speech recognition service (processed 2.5 million hours of audio per day), such as classifying different speakers and limiting quota by accounts

  • Contributed 9k lines of code in Go and 1k to the Go library of department on key-value caching and dumping; reduced over 3 hours of coding on every Go server

• Apple, Machine Learning Engineer Intern (Under NDA), Jan. 2020 - Jun. 2020

  • Proposed a single-stage, light-weight model for image inpainting which was applied to an app

  • Implemented and improved mainstream models in Python to deal with real-life pictures with higher resolution, increasing accuracy of state-of-the-art methods by 3% (in terms of MAE)

  • Designed a novel method to fix blob textures which significantly improved visual experience of bad cases

• Carnegie Mellon University - School of Computer Science, Research Intern, Mar. 2019 - Dec. 2019

  • Proposed an unsupervised pipeline requiring no human annotation for biomedical salient segmentation; published a paper as a first co-author: PUB-SalNet: A Pre-trained Unsupervised Self-Aware Backpropagation Network for Biomedical Salient Segmentation in Algorithms 2020

  • Organized a new biomedical dataset with a wealth of salient objects, different SNR settings and multiple resolutions, named SalSeg-CECT, a resource to conduct pre-training and fine-tuning for other complicated tasks

• Megvii (Face++), Machine Learning Engineer Intern, Jan. 2019 - Dec. 2019

  • Created a light-weight model for depth estimation based on stereo and a single-line Lidar, which overcame severe sparsity of information and was put into service on a delivery robot with a processing time of 0.2s per image on an embedded Jetson TX2 GPU; exploited the classic SGBM algorithm to reduce error at boundaries by 5%

  • Accelerated multi-GPU training by breaking datasets to small objects stored on object storage service (Amazon S3) and pre-downloading onto SSD to save cost of storage by 50%

Publications & Manuscripts

• PUB-SalNet: A Pre-trained Unsupervised Self-Aware Backpropagation Network for Biomedical Salient Segmentation.
  Feiyang Chen*, Ying Jiang*, Xiangrui Zeng and Min Xu. (*equal contribution). Algorithms 2020, 13(5), 126.

• Is Deep Learning All You Need for Unsupervised Saliency Detection of Biomedical Images?
  Feiyang Chen, Ying Jiang, Xiangrui Zeng, and Min Xu. (Manuscript), 2019.

Skills

• Programming languages: C++, Python, Go, Java, Bash, Matlab, Perl

• Backend Development: gRPC, AWS, Redis, Docker, Kafka, Elastic Stack, DevOps, Hadoop, OpenMP, SQL, Django, Linux tools, distributed systems, computer networks, multi-threaded server development & optimization

• ML related: Pytorch, Tensorflow, Keras, PySpark, Computer Vision

Miscellaneous

I have a super cute boyfriend and we have had a wonderful time for 3 years (live update).
I love swimming and have won several champions in my school life. I am also a big fan of TBBT and Sherlock Holmes.