Zhihua Lai (PhD)
š Cambridge, UK | š§ dr.zhihua.lai [AT] gmail.com | š +44 (0)7939 518997 | ILR(UK)
I am currently a Senior Software Engineering in Intelligent Network System, previously known as AFO (Azure For Operators) in the Microsoft Research Cambridge where I contribute and build high performance software (C, C++, eBPF) that run on 5G Base Stations.
During the employment with Amazon, I have contributed to development of several large scale distributed applications in the domain of AWS (Amazon Web Services) e.g. S3 Simple Secure Storage. I have worked from Day 1 at S3 Object Lambda and witnessed the successful launch of it (Mar-18-2021).
During the employment with General Electric, I have added the _while loop, regular expression, http library, jwt APIs and many other features to the Magik compiler (Wiki).
During the employment with Ranplan Wireless, I have designed and implemented the cutting-edge Intelligent Radiowave Propagation Engine that is pushed to its limits by using Win32 inline Assembly in the core. See https://ranplanwireless.com/
In High School (2002), I have won the third prize on NOI programming contest i.e. National Olympiad Informatics.
I am a TOP 3 Witness aka Block Producer of the Steem Blockchain. I am beknown for the contributions (infrastructure, tools, API and services) to the Steem Blockchain. See https://steemyy.com
Microsoft Certified
| Code |
Domain |
Name |
Certification Number |
Earned-on Date |
| AZ-204 |
Cloud |
Azure Developer Associate |
T6D896-FDCEDF |
February 27, 2026 |
| AB-730 |
AI, Business |
AI Business Professional |
BA3490-GD0796 |
February 27, 2026 |
| AI-102 |
AI |
Azure AI Engineer Associate |
25A6FZ-E1AF43 |
February 20, 2026 |
| AZ-900 |
Cloud |
Azure Fundamentals |
B6FAD6-2S7125 |
January 16, 2026 |
| SC-900 |
Security |
Security, Compliance, and Identity Fundamentals |
B6A851-BECO96 |
January 13, 2026 |
| AI-900 |
AI |
Azure AI Fundamentals |
7Z9DA9-388866 |
December 19, 2025 |
| DP-900 |
Data |
Azure Data Fundamentals |
CD4244-E236BG |
December 19, 2025 |
Job Experience
Senior Software Engineer at Microsoft (Cambridge, UK)
2021/07 to Present: Intelligent Network System, Microsoft Research Cambridge
- Full Stack: Linux (BASH), C, C++, eBPF, 4G/5G, O-RAN, Azure (AKS/K8S, Azure Functions, Service Bus, CI/Pipeline ā¦), Python, Go, C#, NodeJS, Rust, Android, WASM etc.
- Role: As an engineer in the Research Org, I support researchers in rapidly prototyping ideas from 0 to 1. Iāve contributed to a wide range of projects, including low-level systems (eBPF in C), performance-critical components in C++, and application development in C#, Android (Java/Kotlin), and robotics (ROS2).
- Open Source:
- Publications:
Software Development Engineer at Amazon AWS (Cambridge, UK)
2020/01 to 2021/07: SDE (L5) = Someone Does Everything = Full Stack Software Engineer + On-Call Duties
- Technology: Java/RxJava, Python, Ruby, Javascript (Node, Vue.Js), AWS S3, EC2, CloudFormation, API Gateway, Lambda, DynamoDB, CloudWatch, Step Function, CDK/LPTā¦
Highlights of contributions:
- Working on S3 Object Lambda from Day 1! https://aws.amazon.com/s3/features/object-lambda/
- Built the internal tooling to track the performance of S3 - the largest distribute storage product in the world. Automated categorization of issues and performance bottlenecks in a very large distributed system help engineers identify and address the problem.
Staff Software Engineer at General Electric (Cambridge, UK)
2018/12 to 2020/01: It has been a tremendous success for me as I have learned a lot, improved myself and become confident in the GE Smallworld technology stacks. I have merged 165+ PRs and completed 92+ stories (25 Pull Requests Merged in First 3 Months), and have contributed quite a few to the smallworld products (Smallworld GIS Software core and other components)
- Technology: BASH, C/C++, Javascript (NodeJS), Java and Magik; REST API, OAuth2 (UAA), Docker, Kubernetes Clusters.
Highlights of contributions:
- I added the while loop to the Magik programming language.
- I added the regular expression to the Magik programming language.
- I designed and implemented the HTTP library.
- I designed and implemented the JWT (Json Web Token) library - used in UAA Authentication.
- I implemented the Magik Random Interop Library that improves the performance and fix the multithreading race condition problems.
- I added the Magik URL validation library in the core.
- I contributed to the development of the outbound/inbound framework.
- I contributed to MUnit testing framework.
Principal Algorithm Engineer at Ranplan Wireless (Cambridgeshire, UK)
2010/09 to 2018/11: I am one of the lead developers for this product (C#7, .NET4.7): https://ranplanwireless.com/products/ and I design, develop and maintain the core library (Radio Propagation engine). It is a 500K LOC code base that consists of Visual Studio C++ 2017, Delphi 10.2 Tokyo with some Inline Assembly and some VBS/JS.
Key result areas included:
- To supportthe development of Ranplan in-house ray tracing model
- To lead cutting-edge channel modelling algorithms and techniques
- To train and share all knowledge of algorithm in radio propagation channel modelling
- To provide level 3 support and customer consulting project services
- To contribute to Ranplan Patents
Highlights:
- Core developer of the flagship product: The Author of Ranplan Radiowave Propagation Model (one of my babies): My overall job purpose was to provide and implement my algorithm research expertise in developing Ranplan Radio Propagation Model (RPM) which is one of Ranplanās key intellectual property.
- Lead a Propagation Team e.g. Rewrite RPM in C++.
- OpenCL GPU Implementation of Vector Ray Launching Propagation Algorithms.
- Prototyping the Geo-localization algorithm using KNN and other machine learning algorithms.
- Develop the 32/64-bit cutting-edge wireless radiowave propagation engine (plugins, scripts) using C++, Delphi and Inline Assembly.
- Research, bug-fixes and development for an indoor wireless planning tool using C#.
- Develop and maintain web-based licensing software using PHP + MySQL + Javascript.
- Develop Hundreds of Unit Tests and Integration Tests in C# and VBScript.
- Develop More than 1500 APIs using Delphi (Win32/64 COM Library).
- Develop Tools of Statistics and Code Quality on Jenkins Continuous Server Powershell.
- Develop Smart/Intelligent algorithms to convert clutter GIS data to vector building data.
- Develop various GIS conversion/cropping tools using C# ie. GeoConverter.
Marie Curie Experienced Researcher at University of Sheffield, UK
2013 to 2015: Development of tools for the design of wireless and energy efficient buildings; in charge of high frequency ray tracing tools. WIFEEB Grant ID: 286333
Awards (Selected)
- 2002: ACM - National Olympic Informatics (NOI), Third Prize by China Computer Federation & NOIP Orgranisation Committee
- 2001: ACM - Fujian Olympic Informatics (FOI), Third Prize by Fujian Education Federation & Fujian Science Orgranisation
Patents (Selected)
- PCT/GB2015/053224, Zhihua Lai etc., Method for Predicting Outdoor Three-Dimensional Space Signal Field Strength By Extended Cost-231-Walfisch-Ikegami Propagation Model, US Patent Application No. 15/522,728, Claimed Filing Date: 27 October 2015.
- PCT/GB2015/053223, Zhihua Lai etc., Method for Predicting Indoor Three-Dimensional Space Signal Field Strength Using An Outdoor-To-Indoor Propagation Model, US Patent Application No. 15/522,735, Claimed Filing Date: 27 October 2015.
Education and Awards
2006-2010 (Three and Half Years)
PhD, Computer Science and Wireless; University of Bedfordshire (UK)
Thesis title: The Development of an Intelligent Ray Launching Algorithm for Wireless Network Planning
- 2009: Marie Curie Fellowship for Transfer of Knowledge by University of Applied Sciences of Western Switzerland
2004-2006
BSc, Computer Science (First Class, Score 15.51 out of 16); University of Luton (UK)
Thesis title: Chinese Chess
- 2004: National Computer Rank Programming Contest, Top 30 by National Education Examinations & China Education
Open Source Projects
I am a builder and a problem-solver.
DAO (Decentralized Autonomous Organization) Proposals
Iāve managed to compile/build the STEEM blockchain (C++) on Modern OS.
Pull Requests:
Selected Publications
- Distributed AI Platform for the 6G RAN, In 2nd ACM Workshop on Open and AI RAN (OpenRan ā25), November 4ā8, 2025, Hong Kong, China. ACM, New York, NY, USA. https://doi.org/10.48550/arXiv.2410.03747
- Accelerating Open RAN Research Through an Enterprise-scale 5G Testbed, In The 29th Annual International Conference on Mobile Computing and Networking (ACM MobiComā23), October 2-6, 2023, Madrid, Spain. ACM, New York, NY, USA. https://doi.org/10.1145/3570361.3615745
- Programmable RAN Platform for Flexible Real-Time Control and Telemetry, In The 29th Annual International Conference on Mobile Computing and Networking (ACM MobiComā23), October 2-6, 2023, Madrid, Spain. ACM, New York, NY, USA. https://doi.org/10.1145/3570361.3614065 - Best Demo Award Runner-Up
- Taking 5G RAN Analytics and Control to a New Level, In The 29th Annual International Conference on Mobile Computing and Networking (ACM MobiComā23), October 2-6, 2023, Madrid, Spain. ACM, New York, NY, USA, 16 pages. https://doi.org/10.1145/3570361.3592493
-
- Chapter 2 - Radio Propagation Modelling, in Book āHeterogeneous Cellular Networks: Theory, Simulation and Deploymentā, Cambridge University Press, 2012.
- Chapter 5 - Outdoor-Indoor Channel, in Book āLTE-Advanced and Next Generation Wireless Networks: Channel Modelling and Propagationā, John Wiley & Sons, November, 2012.
- Intelligent Ray Launching Algorithm for Indoor Scenarios, Radioengineering, Towards EUCAP 2012: Emerging Materials, Methods, and Technologies in Antenna & Propagation, Volume 20, Number 2, 2011, ISSN: 1210-2512, p.p.:398-408.
- The Development of a Parallel Ray Launching Algorithm for Wireless Network Planning, International Journal of Distributed Systems and Technologies, IGI, DOI: 10.4018/jdst.2011040101, Volume 2, Issue 2, 2011.
- Indoor Massive MIMO Channel Modelling Using Ray-Launching Simulation, International Journal of Antennas and Propagation, Volume 2014 (2014), Article ID 279380.
- Joint Ray Launching Method for Outdoor to Indoor Propagation Prediction Based on Interpolation, The Ninth European Conference on Antennas and Propagation, EUCAP, IEEE, Lisbon, Portugal, April 12-17, 2015.
- Joint Ray Launching Method for Indoor to Outdoor Propagation Prediction Based on Ray Aggregation, The Ninth European Conference on Antennas and Propagation, EUCAP, IEEE, Lisbon, Portugal, April 12-17, 2015.
- Implementation and Validation of a New Combined Model for Outdoor to Indoor Radio Coverage Predictions, Hindawi Publishing Corporation EURASIP Journal on Wireless Communications and Networking, 2010: 215352.
- Modelling the mmWave Channel Based on Intelligent Ray Launching Model, The Ninth European Conference on Antennas and Propagation, EUCAP, IEEE, Lisbon, Portugal, April 12-17, 2015.
- Implementation and Validation of a 2.5D Intelligent Ray Launching Algorithm for Large Urban Scenarios, The Sixth European Conference on Antennas and Propagation, EUCAP, IEEE, Prague, Czech Republic, March 26-30, 2012,ISBN: 978-1-4577-0919-7.
- Antenna Height Compensation for an Indoor to Outdoor Channel model based on a 2D Finite Difference Model, 29th Progress In Electromagnetics Research Symposium, Marrakesh, Morocco, March 20-23, 2011.
- A New Deterministic Hybrid Model for Indoor-to-Outdoor Radio Coverage Prediction, The Fifth European Conference on Antennas and Propagation, EUCAP, IEEE, Rome, Italy, April 11-15, 2011, ISBN: 978-88-8202-074-3, p.p.:3771-3774.
Worth mentioning
- Ranked Top 30 iPinyou Global RTB Bidding Algorithm Contest (season two offline, 2013)
- My propagation model āIRLAā has been cited in the book āFemtocells - Technologies and Deploymentā, John Wiley & Sons, ISBN: 978-0-470-74298-3.
- I write highly-optimized code (which uses less memory and executes faster). For example, for Problem 1532 on ACM Timus Online Judge, my submission ranks the second. The maximum time allowed is 4 seconds but my algorithm executes in 0.312 seconds. http://acm.timus.ru/rating.aspx?space=1&num=1532&count=100
- I run a delegated service on steem blockchain \@justyy where Steemians delegate their Steem Power to me and my bots send out the Steem Dollars (tokens) at a rate of 8% to 10%. Also my bots are up running 24/7 to vote the quality posts of the delegatees.
- I am currently the top 3 witness on the STEEM Blockchain. A witness is a block producer.
- I am currently the 3-rd most delegated account on the STEEM Blockchain e.g. 300+ delegations according to https://steemyy.com/top-delegations-by-count/
- I run \@dailychina where the algorithms select top 10 quality posts and reward them daily (very popular and postive feedbacks received).
- I develop \@fairlottery (Blockchain Lottery) and \@witnesstools (Steem Blockchain Gambling).
- I write discord/telegram bots that distribute test coins (TRX, USDT, USDC, USDD) on Tron Blockchain (Shasta and Nile Test Net).
- I develop and run Simple Token Swap Tools on Multiple blockchains as a POC:
- I develop and run Account Registration Services on Steem Blockchain: Register a Free Account on Steem Blockchain.
- I develop the Steem Blockchain Explorer.
- I run a Steem Power Token Rent (Lease Services).
Miscellaneous
- 2019-11 to 2019-12: I taught the course Microbit Programming at weekends in Chesterton Community Colleage (organised by Chinese Family Together and Hacklab Cambridge) to kids (8 yrs+).
- I teach my sons programming (and math) videos (700 days+): https://zhihua-lai.com/teaching/
- Personal Page: https://zhihua-lai.com
- Keep learning:
- Clean UK Driving License (13 yr+)