Franklin Fuchs

Franklin Fuchs

Mondelēz

Al-Folio Migration Package for Franklin Fuchs

Al-Folio Migration Package for Franklin Fuchs

This file contains ALL content and configuration from the old academicpages site, structured for migration to al-folio. Open the cloned al-folio-dev repo in Kiro and paste this file in. Then tell Kiro: “Use MIGRATION_PACKAGE.md to set up my site.”

1. SITE CONFIGURATION

# Key values for _config.yml
first_name: Franklin
last_name: Fuchs
email: fuchs.franklin@gmail.com
location: NYC
employer: Mondelēz
description: >
  Business Intelligence and Forecasting Manager working on optimization, LLM systems, and decision-making under uncertainty.
url: https://fuchsfranklin.github.io
baseurl: ""  # leave empty

# Social
github_username: fuchsfranklin
linkedin_username: franklin-fuchs-aab0a6306

# Theme
theme_color: dark

# No Google Scholar, ORCID, or Twitter currently

2. ABOUT PAGE (Homepage)

Current bio from _pages/about.md:

I am currently working at Mondelēz as a Business Intelligence and Forecasting Manager in the Insights and Analytics Group under Ryan Grum and closely with Matthew Kullmann, where I focus on consumer research and providing product portfolio insights using methods in Algorithmic Marketing, alongside automating analytical workflows and model deployment through cloud computing at scale.

The areas and ideas I (personally) think about a lot often relate to fat tailed distributions, or more accessibly thought of as black swan type phenomena. I also like learning about concepts in psychology, especially behavioural science and mental health. In my free time I like to exercise.

3. CV DATA

Education

  • Johns Hopkins University — ScM in Biostatistics (2021-2023)
  • University of Nevada, Reno — Honors BS in Mathematics (Statistics Emphasis), Minor in Big Data (2017-2021)

Skills

  • Programming Languages: R, Python, SQL, C++, C, MATLAB
  • Software: GCP, AWS, Databricks, BigQuery, Dataiku, Docker, Singularity, SLURM, Markdown, Shiny, LaTeX, Linux Shell, Office Suite, Git, and GitHub
  • Modeling and Computing: Multi-Objective Optimization, Classification, Regression, Time Series Forecasting, Predictive Analytics, Feature Engineering, Variable Selection, Data Wrangling, Exploratory Data Analysis, High-Performance/Cluster Computing, Containerization, Database Design and Implementation, Web-Application Development, Interactive Data Visualization, LLM Infrastructure Testing/Development

Work Experience

  1. Mondelēz Insights and Analytics Group — Business Intelligence and Forecasting Manager (Apr. 2026 - Present)

  2. Mondelēz Insights and Analytics Group — Data Science Manager (Jun. 2025 - Apr. 2026)

  3. Pfizer Digital Client Partners — Data Scientist - DRP Associate (Jun. 2024 - Jun. 2025)
    • Multi-Objective Optimization for target product profile support (Evolutionary Algorithms, RL, Robust Portfolio Optimization)
    • Interactive D3/Python/R visualizations for unmet needs analysis, AWS deployments
    • LLM-enhanced data APIs, RAG systems with source citation, disease evidence map generation
  4. Pfizer AI/ML/Analytics Enterprise Architecture — Software/Solution Engineer - DRP Associate (Jun. 2023 - Jun. 2024)
    • Prototyped Vector Databases and RAG for internal LLM with enterprise scalability (AWS)
    • Enterprise Architecture CityMap tool in Python
    • Evaluation framework documentation for Architecture Data/Analytics Platform
    • Interactive dashboards (Power BI, Tableau, Python, Excel)
  5. Pfizer Machine Learning and Computational Science Group — Data Scientist Intern (Jun. 2022 - Aug. 2022)
    • MySQL database for proteomics data (Chemical Biology Group)
    • R Shiny UI for experiment/data management
    • Extended MSstats R package for proteomics
  6. Pfizer Simulation and Modeling Science Group — Software Development Intern (Jun. 2021 - Aug. 2021)
    • Automated data pipeline in R for computational proteomics cloud (SevenBridges)
    • R Shiny General User Interface
  7. UNR Visual Perception Lab — Student Researcher, HURA Award (Jan. 2021 - May 2021)
    • Sleep classification algorithms (R to MATLAB port)
    • Arduino-based wrist-wearable for circadian rhythm measurement (C/C++)
  8. Nevada INBRE — Student Researcher, INBRE UROP Award (Jan. 2020 - Jan. 2021)
    • R Shiny web-app for TBI prediction model deployment
    • Docker/Singularity containerized parallelized analyses on HPC
    • Pediatric TBI predictive modeling (supervised ML + traditional statistics)
  9. UNR Department of Mathematics — Statistics Research Assistant (Aug. 2019 - Jan. 2020)
    • Bayesian regression analyses for UNR medical school curriculum evaluation
  10. Swiss Armed Forces NBC Defense School — Biology Laboratory Specialist (Jul. 2016 - Dec. 2016)
    • Microbiological hazard identification, BSL-3 laboratory work

Teaching

  • JHU Bloomberg School of Public Health — Teaching Assistant (Aug. 2022 - May 2023)
  • UNR Honors College — Honors Peer Coach and Teaching Assistant (Aug. 2019 - May 2021)
  • UNR Department of Mathematics — Statistics Grader (Jan. 2020 - May 2020)
  • UNR Department of Computer Science — Computer Science Teaching Fellow (Jan. 2018 - Apr. 2019)

Awards

  • Honors Undergraduate Research Award (HURA) — $1,500 + $500 mentor stipend (2021)
  • Nevada INBRE UROP — $6,000 + $1,000 mentor stipend (2020)
  • 1st Place, 2019 Capstone Statistical Computing Project Competition (2019)

4. PUBLICATION (BibTeX)

@article{fuchs2025adiponectin,
  title     = {Association of serum adiponectin and leptin levels with inner retinal thickness among individuals with or without elevated HbA1c},
  journal   = {Scientific Reports},
  year      = {2025},
  url       = {https://www.nature.com/articles/s41598-025-93562-9},
  selected  = {true}
}

Note: The author field was listed as “Co-Authored” on the old site. Franklin should fill in the full author list.

5. PROJECTS (migrate to _projects/)

Project 1: Multi-Agentic Oncology Value Scorecard Creation with LLMs

  • GitHub: https://github.com/fuchsfranklin/multi-agentic-scorecard-creation
  • Description: Replicating established oncology value frameworks (ISPOR Scorecard, ASCO Value Framework) using LLMs. Compares multi-agent systems, single LLM pipelines, and RAG using ClinicalTrials.gov, PubMed, and OpenFDA. A fourth MOA-based multi-agent framework was later integrated.
  • Collaborators: Brett South, Ajit Jadhav, Jay Ronquillo, Jon Mauer, Stephen Watt (Pfizer colleagues)
  • Importance: 1 (flagship)

Project 2: BaltimoreTrails R Package and Dashboard

  • GitHub: https://github.com/datatrail-jhu/BaltimoreTrails
  • Live App: https://fuchsfranklin25.shinyapps.io/shiny-examples/
  • Description: R package and Shiny dashboard for integrating Baltimore datasets into the DataTrail data science education initiative (JHU Bloomberg School of Public Health). ScM thesis work.
  • Importance: 2

Project 3: Survival Analysis Shiny App

  • GitHub: https://github.com/fuchsfranklin/Survival-Analysis-Project
  • Live App: https://franklinf.shinyapps.io/Survival-Analysis-Project/
  • Description: R Shiny web-app introducing survival analysis concepts using a Moderna vaccine dataset. With Tiffany Hsieh and Bowen Chen.
  • Importance: 3

Project 4: Pediatric TBI Mortality Prediction Web-Application

  • GitHub: https://github.com/fuchsfranklin/Pediatric-TBI-Prediction-Application
  • Live App: https://franklinfuchs.shinyapps.io/Pediatric-TBI-Prediction-Application/
  • Description: C5.0 decision tree classifier on SMOTE-subsampled data for pediatric TBI mortality prediction. Part of honors thesis. Published at IEEE BIBM (doi: 10.1109/BIBM49941.2020.9313568).
  • Importance: 2

Project 5: Metropolis-Hastings Algorithm Visualization

  • GitHub: https://github.com/fuchsfranklin/MCMC-Visualization-Project
  • Live App: https://franklinfuchs.shinyapps.io/MCMC_Visual_Project/
  • Description: Interactive animated visualizations of the Metropolis-Hastings algorithm and MCMC diagnostics in R Shiny.
  • Importance: 4

Project 6: Variable Selection Techniques on Simulated Data

  • GitHub: https://github.com/fuchsfranklin/Regularization-Project
  • Live: https://rpubs.com/franklinfuchs/Regularization-Project
  • Description: Simulating multivariate data from Tibshirani (1996) Lasso paper and comparing variable selection methods (AIC, BIC, Adjusted R²).
  • Importance: 5

Project 7 (HIDDEN on old site): COVID-19 Modeling

  • GitHub: https://github.com/fuchsfranklin/Exploratory-Covid-Modeling
  • Description: COVID-19 pandemic dynamics through healthcare strain, pandemic fatigue, and policy effectiveness. Ensemble methods for ICU utilization prediction.
  • Importance: 6

SKIP: Old Shiny Personal Website (portfolio-6) — retired, don’t migrate

6. BLOG POSTS (migrate to _posts/)

Post 1: Pure Indexing, Factor Tilts, Three-Fund Portfolio

  • Date: 2025-12-27
  • Old permalink: /posts/2025/12/pure-indexing-factor-tilts-three-fund-portfolio/
  • Tags: investing, index-funds, factor-investing, bogleheads, personal-finance
  • Images needed: vti_vxus_mix_growth.png, vti_dfus_growth.png, vxus_dfax_growth.png
  • Code files needed: vti_vxus_mix_comparison.py, vti_dfus_comparison.py, vxus_dfax_comparison.py, factor_regression_analysis.py
  • External code repo: https://github.com/fuchsfranklin/balancing-priorities-project
  • Note: Contains custom CSS for larger code blocks and images — will need al-folio equivalent or can be dropped (al-folio handles this natively)

Post 2: Breaking Deep Research — Where LLM Search Agents Fail

  • Date: 2026-03-03
  • Old permalink: /posts/2026/03/breaking-deep-research-where-llm-search-agents-fail/
  • Tags: AI, deep-research, LLMs, hallucination, verification
  • Images needed: none (tables only)
  • Note: Contains HTML <details> blocks for appendix — al-folio supports these natively. Also has custom CSS that can be dropped.

7. IMAGES TO COPY

From old repo images/ → new repo assets/img/:

  • profile.png → prof_pic.jpg (or keep as .png, al-folio supports both)
  • vti_vxus_mix_growth.png
  • vti_dfus_growth.png
  • vxus_dfax_growth.png
  • tbi_prediction_webapp.png

8. FILES TO COPY

From old repo files/ → new repo assets/:

  • files/blog/factor_regression_analysis.py → assets/code/factor_regression_analysis.py
  • files/blog/vti_dfus_comparison.py → assets/code/vti_dfus_comparison.py
  • files/blog/vti_vxus_mix_comparison.py → assets/code/vti_vxus_mix_comparison.py
  • files/blog/vxus_dfax_comparison.py → assets/code/vxus_dfax_comparison.py
  • files/paper1.pdf, paper2.pdf, paper3.pdf → check if these are real papers or template files

9. NAVIGATION (old site had 3 items)

Old nav:

  1. Expanded Resume/CV → /cv/
  2. Projects → /portfolio/
  3. Data & Daydreams (Blog) → /year-archive/

Suggested al-folio nav (configured in _data/navigation.yml or _config.yml):

  1. About (homepage, default)
  2. Projects
  3. Publications
  4. Blog (keep “Data & Daydreams” as subtitle or description)
  5. CV

10. POSTERS/PRESENTATIONS (from CV, for reference)

These are listed in the CV but don’t have dedicated pages:

  • “Association of serum adiponectin…” — Scientific Reports 2025 (PUBLICATION, handled above)
  • “Enhancing DataTrail…” — ScM Thesis (2023) (PROJECT, handled above)
  • “ProtIOT” — Pfizer MLCS group talk (2022)
  • “Standardized Proteomics Pipeline” — Pfizer SMS/MI demos (2021)
  • “Low-Cost Arduino-Based Wearable” — UNR Symposium (2021)
  • “Pediatric TBI Survival Prediction” — IEEE BIBM + UNLV Symposium (2020) (doi: 10.1109/BIBM49941.2020.9313568)
  • “Intuitive Introduction to Metropolis-Hastings” — Personal Project (2020)
  • “Bayesian Regression Model” — UNR Capstone Competition (2019)
  • “Comparing Regularization Techniques” — Personal Project (2019)

11. DARK THEME SETUP

In al-folio’s _config.yml, set:

# Theme
theme_color: dark

This enables the dark theme globally. Users can toggle between light/dark via the sun/moon icon in the navbar.

To default to dark and still allow toggling, this is the default behavior when theme_color: dark is set.

12. LOCAL DEVELOPMENT

# Option A: Docker (recommended)
docker compose pull
docker compose up
# Visit http://localhost:8080

# Option B: VS Code Dev Container
# Just open the repo in VS Code, accept the Dev Container prompt

# Option C: Native Ruby (if you have it)
bundle install
bundle exec jekyll serve
# Visit http://localhost:4000

13. DEPLOYMENT CHECKLIST (when ready to go live)

  1. Rename current fuchsfranklin.github.io repo → old-site-backup
  2. Rename al-folio-dev → fuchsfranklin.github.io
  3. In _config.yml, confirm url: https://fuchsfranklin.github.io and baseurl: ""
  4. In repo Settings → Actions → General → Workflow permissions → Read and write
  5. Push to main branch
  6. Wait for “Deploy site” GitHub Action (~4 min)
  7. In repo Settings → Pages → set branch to gh-pages
  8. Wait for pages-build-deployment action (~45s)
  9. Visit https://fuchsfranklin.github.io