Research Interests

Humanoid AI: Reinforcement Learning, Simulation to Reality, Inference Time Scaling, Post-Processing Action.

Deep Learning: AI Coding Assistant, Code LLM, Multi-Node Distributed Training, Parameter Efficient Fine-Tuning, Instruction Tuning, LLM Inference Server, Prompt Engineering, Benchmark Dataset, Data Collection and Cleaning, Time Series Forecasting, Semantic Parsing.

Algorithm Engineering: Fast and Scalable Algorithms, Graph Isomorphism, Subgraph Matching, Multiple String Matching, Cartesian Tree Matching, Order-Preserving Matching, Traveling Salesman Problem, Approximating Polygons and Subdivisions with Minimum-Link Paths, Path Simplification.

Work Experience

LG Electronics - Artificial Intelligence Lab (Senior Researcher)

• Jan. 2025 - Present: Humanoid AI
• Jan. 2024 - Dec. 2024: Development of AI Coding Assistant using Large Language Model
  • Conducted research on continual pretraining code LLMs in data scarce scenario.
  • Maintained custom benchmark dataset for offline evaluation.
  • Analyzed user statistics and feedback for online evaluation.
  • Constructed instruction dataset and conducting instruction-tuning.
  • Prompt engineering for accurate code suggestion.
• Aug. 2022 – Dec. 2023: Development of AI Coding Assistant using Large Language Model
  • Conducted distributed training of LLMs based on decoder-only transformer on AWS.
  • Filtered and deduplicated terabytes of source code data.
  • Conducted research about data augmentation, forgetting, and efficient fine-tuning of LLM.
  • Developed a fast LLM inference server based on NVIDIA Triton and FasterTransformer.
• Apr. 2022 – Dec. 2022: Development of Coding Education Program Utilizing AI
  • Constructed training data for generating Python code from natural language instruction.
  • Trained an encoder-decoder transformer from scratch.
  • Developed a web client that inputs prompt, prints AI-generated code, and executes Python code.
  • Created an inference server that runs on multiple GPUs, loads multiple copies of the model, and offers dynamic batching for increased throughput.

Seoul National University – Institute of Computer Technology (Post-Doctoral Assistant)

• Sept. 2021 and Jan. 2022 – Mar. 2022: Algorithm Development for Graph Isomorphism Query Processing
  • Developed a fast graph isomorphism query processing algorithm that runs orders of magnitude faster than state-of-the-art algorithms.

NAVER – AI Dev2 (Internship)

• Oct. 2021: Analyzing Conversion Tracking Data
  • Conducted exploratory data analysis on glad for advertisement data to find meaningful trends.
  • Handled hundred gigabytes of (raw) conversion tracking data.
  • Solved optimization problem of maximizing conversion rate using linear programming.

Skills

Programming Languages

• C/C++, Python, CUDA C++, Rust, C#, Java, Shell Script, LaTeX

Libraries

• PyTorch, TensorFlow, HuggingFace Transformers, DeepSpeed, Triton (NVIDIA), FasterTransformer, FastAPI, Triton (OpenAI), Seaborn, Pandas, PySpark, gtest

Competitive Programming

Framework

• AWS (SageMaker, EC2, Lustre, S3), Docker, ROS 2

CV

GeonmoGu_CV