Skip to content
View kyb8801's full-sized avatar

Block or report kyb8801

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
kyb8801/README.md

Hi, I'm Yongbeom Kim 👋

Metrology × AI × Uncertainty Quantification

Optics Ph.D. with 8+ years in optical & semiconductor metrology (SEM, AFM, Raman, NSOM), building AI applications that don't just predict — they report how much you can trust the prediction, with GUM-compliant uncertainty budgets (JCGM 100/101).

  • 🔬 Measurement scientist first: ISO 18516 round-robin lead contributor · KOLAS-accredited calibration & testing background
  • 📐 I build the full loop: physics forward model → inverse solver → GUM uncertainty budget
  • 🧰 Domains: OCD / scatterometry, XRR, SEM·TEM image analysis, AFM probe health, Raman/SERS QC
  • 🤖 MCP builder: author of the first Model Context Protocol server for GUM measurement uncertainty

🚀 Featured Projects

Project What it does Stack
metrology-inverse Forward → inverse → GUM uncertainty across 3 instruments: OCD (RCWA), XRR (Parratt), autodiff CD fitting — validated on real NIST scatterometry data (L100P300, 9 dies). Exact-Jacobian sensitivity for the uncertainty budget. Python, PyTorch, Meent, refnx
measurement-uncertainty-mcp First MCP server for GUM-compliant uncertainty analysis: Type A/B, Welch–Satterthwaite ν_eff, expanded U(k), JCGM 101 Monte Carlo, KOLAS-ready budgets. Python, MCP
tiphealth Recipe-aware HAR AFM tip predictive maintenance — hybrid Archard physics + ML, 8 industrial probes × 6 materials, NIST-calibrated wear rates, conformal prediction intervals, Dockerized API. Python, Docker, Streamlit
spectraguard Uncertainty-aware spectral QC for SERS/Raman/IR — 6-metric confidence score with bootstrap CIs, cross-instrument transfer, streaming SPC, CLI + CI pipeline. Python, NumPy, SciPy
semiconductor-defect-classifier Defect classification on SECOM fab sensor data (1,567 wafers × 590 sensors, 6.6% defect rate) — imbalance handling, Optuna tuning, honest K-fold evaluation. Python, XGBoost, Optuna
semiconductor-ai-portfolio Analysis pipelines on my own PhD measurement data: MoSe₂ photoluminescence peak/FWHM analysis, NSOM defect mapping, TMD comparison. Jupyter, pandas, SciPy

🔌 MCP & Open Source

Model Context Protocol servers I built and maintain:

🛠️ Tech

Python PyTorch XGBoost Optuna NumPy SciPy pandas Jupyter Streamlit Docker GitHub Actions

Metrology & standards: GUM (JCGM 100:2008) · JCGM 101 Monte Carlo · ISO/IEC 17025 · RCWA · XRR · SEM/TEM · AFM · Raman/SERS · NSOM

📫 Contact

Popular repositories Loading

  1. semiconductor-ai-portfolio semiconductor-ai-portfolio Public

    AI for semiconductor metrology | Optics PhD portfolio

    Jupyter Notebook 1

  2. semiconductor-defect-classifier semiconductor-defect-classifier Public

    Deep learning model for semiconductor defect classification using SEM/TEM images

    Jupyter Notebook

  3. spectraguard spectraguard Public

    Uncertainty-aware spectral quality assessment for SERS, Raman, and IR spectroscopy. 6-metric confidence scoring with bootstrap CI.

    Python

  4. metroai metroai Public

    KOLAS Compliance OS — MCP server + 6 AI agents + GUM/MCM/QMC measurement uncertainty engine + Ed25519-signed audit trail. For ISO/IEC 17025 accredited labs.

    Python

  5. measurement-uncertainty-mcp measurement-uncertainty-mcp Public

    The first Model Context Protocol server for GUM-compliant measurement uncertainty analysis

    Python

  6. awesome-mcp-servers awesome-mcp-servers Public

    Forked from punkpeye/awesome-mcp-servers

    A collection of MCP servers.