update resume from my computer

Author: alegresor

resume/sorokin_resume.pdf

@@ -1,12 +1,12 @@
 \documentclass[a4paper, sans, 11pt]{moderncv}
 
+\moderncvcolor{blue}
 \moderncvstyle{banking}
-\moderncvcolor{black}
 
 \usepackage{multicol}
 \usepackage{lipsum}
 \usepackage[hyperref]{}
-\usepackage[scale=0.8, top=.5cm, bottom=.5cm, left=.8cm, right=.8cm]{geometry}
+\usepackage[scale=0.8, top=.5cm, bottom=.5cm, left=.725cm, right=.725cm]{geometry}
 
 \usepackage{fontawesome5}
 \usepackage{lmodern}
@@ -64,17 +64,17 @@ \subsection{Experiences}
 \newentry{\normalfont{2025.01 - 2025.12}}{\textbf{DOE SCGSR Fellow in Applied Math} at \textbf{Sandia National Laboratory} in Livermore, CA. I produced scientific ML models for machine-precision solutions to nonlinear PDEs \cite{bacho.CHONKNORIS}. I developed scalable multi-fidelity Gaussian processes regression models and open-source software implementations \cite{sorokin.FastBayesianMLQMC,sorokin.fastgps_probnum25}.}
 \newentry{\normalfont{2024.05 - 2024.08}}{\textbf{Scientific Machine Learning Researcher} at \textbf{FM (Factory Mutual Insurance Company).} I deployed scientific ML models, including PINNs DeepONets, to accelerate CFD fire dynamics simulations \cite{sorokin.RTE_DeepONet}.}
 \newentry{\normalfont{2023.05 - 2023.08}}{\textbf{Graduate Intern} at \textbf{Los Alamos National Laboratory.} I modeled multi-fidelity solutions to PDE with random coefficients using efficient and error aware Gaussian processes regression models \cite{sorokin.gp4darcy}.}
-\newentry{\normalfont{2022.05 - 2022.08}}{\textbf{Givens Associate Intern} at \textbf{Argonne National Laboratory}. I derived error bounds and proposed a sequential sampling method for efficiently estimating failure probabilities with probabilistic models \cite{sorokin.adaptive_prob_failure_GP}.}
+\newentry{\normalfont{2022.05 - 2022.08}}{\textbf{Givens Associate Intern} at \textbf{Argonne National Laboratory}. I derived error bounds and a sequential sampling method for efficiently estimating failure probabilities with probabilistic models \cite{sorokin.adaptive_prob_failure_GP}.}
 \newentry{\normalfont{2021.05 - 2021.08}}{\textbf{ML Engineer Intern} at \textbf{SigOpt, an Intel Company}. In a six-person ML team, I contributed production code for meta-learning model-aware hyperparameter tuning via Bayesian optimization \cite{sorokin.sigopt_mulch}.}
 \newentry{\normalfont{2022.09 - 2022.11}}{\textbf{Participant} in \textbf{Argonne National Laboratory's Course on AI Driven Science on Supercomputers}. Key topics included handling large scale data pipelines and parallel training for neural networks.} %\itlink{github.com/alegresor/ai-science-training-series}{https://github.com/alegresor/ai-science-training-series}.
 \newentry{\normalfont{2018.05 - 2019.08}}{\textbf{Instructor} for the \textbf{STARS Computing Corps' Computer Discover Program.} I taught and developed curriculum for middle school and high school girls to learn programmatic thinking in Python.}
 \newentry{\normalfont{2021.08 - 2025.01}}{\textbf{Teaching Assistant} at \textbf{IIT}. I led reviews for PhD qualifying exams in analysis and computational math.}
 
 \subsection{Open-Source Software}
-\newentry{\texttt{QMCPy}}{\textbf{Quasi-Monte Carlo Python Software} (\href{https://qmcsoftware.github.io/QMCSoftware}{qmcsoftware.github.io/QMCSoftware}). I led dozens of collaborators across academia and industry to develop QMC sequence generators, automatic variable transformations, adaptive error estimation algorithms, and diverse use cases \cite{sorokin.thesis,sorokin.2025.ld_randomizations_ho_nets_fast_kernel_mats,choi.challenges_great_qmc_software,choi.QMC_software,sorokin.MC_vector_functions_integrals,sorokin.QMC_IS_QMCPy,hickernell.qmc_what_why_how,jain.bernstein_betting_confidence_intervals}.}
+\newentry{\texttt{QMCPy}}{\textbf{Quasi-Monte Carlo Python Software} (\href{https://qmcsoftware.github.io/QMCSoftware}{qmcsoftware.github.io/QMCSoftware}). I led dozens of collaborators across academia and industry to develop QMC sequence generators, automatic variable transformations, adaptive error estimators, and diverse use cases \cite{sorokin.thesis,sorokin.2025.ld_randomizations_ho_nets_fast_kernel_mats,choi.challenges_great_qmc_software,choi.QMC_software,sorokin.MC_vector_functions_integrals,sorokin.QMC_IS_QMCPy,hickernell.qmc_what_why_how,jain.bernstein_betting_confidence_intervals}.}
 \newentry{\texttt{FastGPs}}{\textbf{Scalable Gaussian Processes in Python} (\href{https://alegresor.github.io/fastgps}{alegresor.github.io/fastgps}). This supports GPU scaling, batched inference, hyperparameter optimization, multi-fidelity GPs, and efficient Bayesian cubature. \texttt{FastGPs} is the first package to implement GPs which require only $\mathcal{O}(n)$ storage and $\mathcal{O}(n \log n)$ computations compared to the typical  $\mathcal{O}(n^2)$ storage and $\mathcal{O}(n^3)$ computations requirements \cite{sorokin.fastgps_probnum25,sorokin.FastBayesianMLQMC}.}
 \newentry{\scalebox{.9}{\texttt{QMCGenerators.jl}}}{\textbf{Randomized Quasi-Monte Carlo Sequences in Julia} (\href{https://alegresor.github.io/QMCGenerators.jl}{alegresor.github.io/QMCGenerators.jl}).}
-\newentry{\texttt{QMCToolsCL}}{\textbf{Randomized Quasi-Monte Carlo Sequences in C / OpenCL} (\href{https://qmcsoftware.github.io/QMCToolsCL/}{qmcsoftware.github.io/QMCToolsCL/}).}
+\newentry{\texttt{QMCToolsCL}}{\textbf{Randomized Quasi-Monte Carlo Sequences in C/OpenCL} (\href{https://qmcsoftware.github.io/QMCToolsCL/}{qmcsoftware.github.io/QMCToolsCL/}).}
 \newentry{\scalebox{.95}{\texttt{TorchOrthoPolys}}}{\textbf{Orthogonal Polynomials in PyTorch} (\href{https://alegresor.github.io/TorchOrthoPolys/}{alegresor.github.io/TorchOrthoPolys/}) with GPU support.}
 
 \subsection{Awards}