resume
Alejandro Gomez Espinosa Ph.D.
alefisicosocial@gmail.com. Personal website. Linkedin profile. Currently in Quito, Ecuador.
Lead Research Scientist and Quantitative Modeler with over a decade of experience designing and deploying machine learning solutions across petabyte-scale datasets at CERN and Carnegie Mellon University. Expert in developing novel statistical estimation algorithms, deep learning models, and synthetic data techniques to extract rare signals in high-noise, high-uncertainty environments. Proven track record architecting distributed computing pipelines and declarative workflow orchestrators (Kubernetes, Slurm, CI/CD) engineered to operate seamlessly with agentic AI systems. Demonstrated technical and cross-functional leadership, coordinating international teams of 20+ researchers and delivering high-stakes findings to global stakeholder audiences of over 500.
Skills
- Programming languages: Python, C++, Bash, SQL, R, LaTeX.
- Technical tools: Pandas, Numpy, Scipy, Jupyter-notebooks, Scikit-learn, TensorFlow, Bigquery, Matplotlib, Seaborn, Docker, Kubernetes, Tidyverse, ggplot2, Linux, Git, Spreadsheets, Tableau.
- Machine Learning: Regression, Classification, Feature engineering, Data Scraping, MC Methods.
- Soft skills: Lead and collaborate with multicultural groups of 20+ people in person and remotely. Communicate results to 500+ stakeholders in conferences and write scientific peer-review documents.
- Languages: English (fluent), Spanish (Native).
Professional Experience
Research Physicist, Carnegie Mellon University (2026 – Present)
- Leading the CMS Run 3 inclusive ttH(bb) analysis, spearheading a modernized analysis strategy that substantially outperforms previous measurements.
- Co-leading CMS Run 3 di-Higgs searches in the bbbb final state, advancing data-driven background modeling and uncertainty frameworks to maximize search sensitivity.
- Developing machine learning algorithms for signal classification and out-of-sample density ratio estimation.
- Co-supervising graduate researchers (Master’s and Ph.D.) and administering the local research cluster (Slurm, GPUs).
Postdoctoral Research Associate, Carnegie Mellon University (2023 – 2026)
- Co-led the CMS Run 2 di-Higgs to four b-quark (bbbb) analysis group, improving sensitivity by 30% using data-driven background estimations.
- Pioneered low-pT jet reconstruction strategies in di-Higgs searches, boosting overall sensitivity by 10% (bbbb) and ~20% (bbWW).
- Fully preserved the 120+ job analysis workflow using Snakemake and REANA, and built multi-stage CI/CD pipelines (50+ jobs).
- Managed local group cluster resources (Slurm queues, GPU nodes) and led community tutorials on workflow orchestration.
Data Scientist, CERN-CMS Experiment (2011 – Present)
- Built an academic career over a decade as part of a team of 2000+ stakeholders in one of the CERN experiments, representing three different institutions.
- Designed and optimized over five multi-class C++/Python frameworks to analyze petabyte-scale data and simulations, resulting in significant acceleration of processing times from weeks to days.
- Authored and co-authored over five peer-reviewed scientific publications.
- Leadership Experience:
- Chair, U.S. CMS Mentorship Committee (2026 – Present; Member 2024–2026) - Directing national mentorship pairings and professional development initiatives across member institutions.
- Workflow Orchestration and Analysis Preservation Convener (2024-2026) - Directed the CMS Experiment efforts on workflow orchestration tools, CI/CD practices, and analysis preservation.
- Jet Algorithms Convener (2019-2021) - Directed and led a research group of over 20 stakeholders, successfully coordinating with competing partners to produce and publish reports.
- Trigger Subgroup Convener (2016-2017). Coordinated the deployment of real-time data filters for a research group consisting of more than 30 stakeholders.
Postdoctoral Research Assistant, ETH Zurich (2018 – 2022)
- Managed the development of two machine learning-based techniques, which were trained in simulations and implemented in data, with the aim of enhancing the filtration of rare theoretically predicted phenomena. The successful implementation of these techniques resulted in a 10% increase in filtering efficiency.
- Recognized and investigated two underexplored research areas and developed innovative strategies to address them, resulting in the publication of scientific research papers.
- Oversaw the completion of three Ph.D. and two master’s-level final projects as a supervisor.
Research Assistant, Rutgers - The State University of New Jersey (2012 – 2018)
- Conducted feasibility studies of novel theoretical models and implemented new data-driven techniques to estimate predictions, resulting in the authorship of two scientific publications.
- Optimized signal-to-background discriminant variables, achieving a 20% improvement over previous results.
- Conceptualized, designed, and deployed three real-time data collection filters, optimizing the probability of capturing scarce yet significant data for the first time.
Visitor Research Student, Fermilab (2011 – 2012)
- Implemented a novel analysis framework using C++ and multivariate techniques, specifically boosted decision trees, to study the production of rare theoretical models for the first time.
- The results obtained during this internship program were incorporated into my bachelor’s project and contributed to a published research paper.
Education
- Doctor of Philosophy, Experimental Particle Physics. Rutgets, The State University of New Jersey - USA. (2012-2018)
- Bachelor of Science, Physics. Escuela Politecnica Nacional - Ecuador (2007-2012)
- Continuous Education:
- Google Data Analytics Professional Certificate - Google (Coursera, 2022)
- Deep Learning Specialization - Deep Learning.AI (Coursera, 2022)
- Sport Performance Analytics Specialization - U. Michigan (Coursera, 2021)
Additional information
As an Ecuadorian citizen, I require support with obtaining a work visa for relevant job opportunities.