🚀 HPC Performance Engineer | AI Systems Builder | Cybersecurity Engineer
🎓 Final-Year CS Student @ Nanyang Technological University Singapore

I work at the intersection of high-performance computing, AI infrastructure, and security engineering — optimizing large-scale systems, profiling real workloads, and building automation that actually survives production environments.

• Performance analysis and optimization of GROMACS, NWChem, SeisSol, SST
• CPU/GPU benchmarking on V100, A100, L40S, H100, H200 and AMD EPYC / Intel platforms
• Multi-node MPI scaling, energy efficiency analysis (Intel RAPL), and IO benchmarking
• Profiling with Arm Forge (MAP/DDT), IPM, NVIDIA Nsight Systems
• Ran and optimized workloads on NSCC Aspire2A/2A+, NCI Gadi, PSC Bridge2

• Fine-tuned LLaMA 3.1 (8B & 70B) using Megatron-LM + PyTorch on DGX H100
• Optimized multi-GPU training throughput and NCCL communication
• Built agentic AI systems using LangGraph for workflow automation
• Experience with LitGPT, Ollama (Qwen3), MLPerf inference

• SIEM rule tuning, alert correlation, and forensic log analysis
• Vulnerability management, penetration testing, and IR playbook authoring
• Built SOC automation with Wazuh, TheHive, Shuffle SOAR
• Developed AI-assisted vulnerability verification to reduce false positives

• 🥈 2nd Overall & Best AI Performance — 7th APAC HPC-AI Competition
• 🥉 3rd Overall — 8th APAC HPC-AI Competition
• 🏅 Top 4 Overall - ISC25 Student Cluster Competition
• 🏅 Top 4 Overall - SC25 Student Cluster Competition
• 🎖 Top 10 Finalist — TechFest 2025

Python | FastAPI | LangGraph | CodeQL | React | Docker

• Reduced CodeQL false positives by 85% using multi-agent AI verification
• Built 5-node LangGraph system for static analysis, classification & remediation
• Real-time WebSocket UI with explainable AI reasoning

• Optimized LLaMA-70B training & inference on national supercomputers
• SeisSol earthquake simulation & 3D visualization (Turkey 2023 dataset)
• NWChem DFT energy simulations with performance/energy trade-off analysis
• IO500, HPL, HPL-MXP benchmarking on in-house and national clusters

• 3-node Supermicro cluster (576 CPU cores, 20× H100 80GB GPUs)
• 400 Gbps ConnectX-7 InfiniBand
• Prometheus + Grafana monitoring, automated alerting, thermal tracking

HPC & Performance
MPI CUDA Nsight Systems Arm Forge IPM Intel RAPL HPL IO500

AI / ML
PyTorch Megatron-LM LangGraph LitGPT Ollama LLMs

Security
Wazuh TheHive Shuffle SOAR Nessus Metasploit Wireshark

Infra & Dev
Linux Docker Kubernetes AWS EKS FastAPI React PostgreSQL

• 💼 Interested in HPC research, AI infra, or security engineering
• 🤝 Open to collaborations, competitions, and research projects

📧 Email: yoongkentan47@gmail.com 💼 LinkedIn: [linkedin.com/in/tan-yoong-ken-972a65252]

I enjoy turning complex systems into measurable, scalable, and efficient ones.