Siddhartha Srinivasa CV

This repository contains the LaTeX source for Siddhartha Srinivasa's academic CV with automated YAML-to-CSV conversion and cross-repository build triggers.

Quick Start

# Using Makefile (recommended)
make all              # Sync data + pubs + build CV
make sync-data        # Fetch latest YAML and generate CSVs
make sync-pubs        # Fetch latest publications
make build            # Build CV PDF
make clean            # Clean auxiliary files
make test             # Validate YAML data

# Or use individual commands
python3 yaml2csv.py --owner siddhss5 --repo siddhss5.github.io --branch main --output-dir data/
latexmk -pdf sidd-cv.tex

Features

• YAML Source of Truth: People from the lab's pubs repo; awards and press from the website repo
• Automated CSV Generation: Python script fetches YAML and generates sorted CSV files
• Cross-Repo Automation: CV automatically rebuilds when those YAML files change
• Smart Sorting: Built-in sorting eliminates need for manual data ordering
• Full Bibliography Support: Handles journals, conferences, and miscellaneous publications
• URL Support: Clickable links in press coverage section

File Structure

├── sidd-cv.tex              # Main CV document
├── yaml2csv.py              # Script to convert YAML → CSV
├── latexmkrc                # latexmk configuration
├── data/                    # CSV data files (generated from YAML)
│   ├── students-phd.csv     # PhD students (generated)
│   ├── students-ms.csv      # MS students (generated)
│   ├── postdocs.csv         # Postdoctoral fellows (generated)
│   ├── grants.csv           # Research grants (hand-edited)
│   ├── awards.csv           # Awards (generated)
│   └── press.csv            # Press coverage (generated)
├── pubs/                    # Bibliography files
│   ├── siddpubs-journal.bib
│   ├── siddpubs-conf.bib
│   └── siddpubs-misc.bib
└── *.tex                    # Section files (mentoring, press, etc.)

Data Management

YAML Source of Truth

Most CV data is maintained in YAML files elsewhere:

• people.yaml in personalrobotics/pubs - PhD students, MS students and postdocs, including the roles a person held before their current or last one (earlier_roles). The lab website and goodrobot.ai read the same file.
• data/awards.yaml in the website repository - Awards and honors
• data/press.yaml in the website repository - Press coverage

Exception: data/grants.csv remains hand-edited in this repository.

Generating CSV Files

Run yaml2csv.py to fetch the YAML and generate CSV files (--people-owner, --people-repo and --people-branch say where people.yaml is, defaulting to personalrobotics/pubs on master):

python3 yaml2csv.py --owner siddhss5 --repo siddhss5.github.io --branch main --output-dir data/

The script automatically:

• Fetches the latest YAML, and fails if people.yaml cannot be fetched rather than keep stale mentoring tables
• Converts YAML to CSV format that LaTeX datatool expects
• Applies proper sorting (e.g., PhD students by Finish:desc, Start:desc)
• Handles current vs. alumni distinctions (empty end_year fields)

Automated Updates

This repo → the website. When a build commits a rebuilt sidd-cv.pdf, it sends a cv-updated repository_dispatch to siddhss5/siddhss5.github.io, which rebuilds and fetches the PDF into site/assets/. The website no longer keeps its own copy. This needs a fine-grained PAT with Contents: write on the website repo, stored here as the secret SITE_REPO_PAT; without it the dispatch step fails on its own and the website picks the CV up on its next build instead.

The website → this repo. When awards.yaml or press.yaml changes there, the website's trigger-cv-build.yml dispatches yaml-updated here and build.yml rebuilds. That needs a PAT with Actions: write on this repo, held in the website repo as CV_REPO_PAT.

personalrobotics/pubs → this repo. When people.yaml or a .bib file changes on pubs' master, its notify.yaml dispatches yaml-updated here the same way, with the same kind of PAT held in pubs as CV_REPO_PAT.

The daily backstop. sync-pubs.yml refreshes pubs/*.bib from personalrobotics/pubs, regenerates the CSVs, and rebuilds if either changed. It runs daily, so anything the two dispatches miss — a paper award added to a .bib file, for instance, which nothing notifies about — is picked up within a day. It used to run weekly and to check only the .bib files, so a change to the website's YAML alone rebuilt nothing at all.

Compilation

Using latexmk (Recommended)

# Build the CV
latexmk -pdf sidd-cv.tex

# Force rebuild (useful when CSV files change)
latexmk -pdf -g sidd-cv.tex

# Clean auxiliary files
latexmk -c

Manual Compilation

# Generate CSV files from YAML first
python3 yaml2csv.py --owner siddhss5 --repo siddhss5.github.io --branch main --output-dir data/

# Compile LaTeX
pdflatex sidd-cv.tex
bibtex jour
bibtex conf
bibtex misc
pdflatex sidd-cv.tex
pdflatex sidd-cv.tex

Configuration

latexmkrc

The latexmkrc file configures the build process:

• Monitors CSV files for changes
• Uses pdflatex as the default engine
• Cleans auxiliary files automatically

yaml2csv.py

The Python script handles YAML-to-CSV conversion:

• Fetches YAML files from website repo via GitHub API
• Maps YAML fields to CSV columns (e.g., co_advisor → Coadvisor)
• Applies sorting rules for each data type
• Handles empty fields correctly (e.g., current students have no end_year)

Dependencies

• LaTeX: TeX Live 2025 or later
• Python 3.9+: For yaml2csv.py script
• PyYAML: Install with pip install pyyaml
• latexmk: For automated compilation
• datatool: LaTeX package for CSV processing

Troubleshooting

CSV Generation Issues

If CSV files aren't being generated:

# Check if yaml2csv.py runs
python3 yaml2csv.py --owner siddhss5 --repo siddhss5.github.io --branch main --output-dir data/

# Verify PyYAML is installed
pip install pyyaml

# Force rebuild with latexmk
latexmk -pdf -g sidd-cv.tex

LaTeX Compilation Issues

If compilation fails:

# Clean all auxiliary files
latexmk -c

# Check for missing dependencies
tlmgr install datatool

Bibliography Issues

If references aren't resolved:

# Force bibliography rebuild
latexmk -pdf -g sidd-cv.tex

Development

Adding New Data

1. For people, awards, press: Edit YAML files in the website repo
   • Push changes to trigger automatic CV rebuild via repository_dispatch
2. For grants: Edit data/grants.csv directly in this repo (hand-edited)
3. Run latexmk -pdf sidd-cv.tex to rebuild locally

Modifying Sort Order

Edit the SORT_SPECS dictionary in yaml2csv.py:

SORT_SPECS = {
    "students-phd.csv": [("Finish", "desc"), ("Start", "desc")],
    # Add or modify sort rules here
}

Local Development

Python Environment Setup

For development, it's recommended to use a virtual environment:

# Create virtual environment
python3 -m venv venv

# Activate virtual environment
source venv/bin/activate  # On macOS/Linux
# or: venv\Scripts\activate  # On Windows

# Install dependencies
pip install pyyaml

Testing Changes Locally

Before committing, test your changes:

# 1. Update CSV files from YAML
python3 yaml2csv.py --owner siddhss5 --repo siddhss5.github.io --branch main --output-dir data/

# 2. Compile the CV
latexmk -pdf sidd-cv.tex

# 3. Verify the PDF output
open sidd-cv.pdf  # macOS
# or: xdg-open sidd-cv.pdf  # Linux

Managing Publications

The pubs/ directory contains BibTeX files for different publication types:

• siddpubs-journal.bib - Journal papers
• siddpubs-conf.bib - Conference papers
• siddpubs-misc.bib - Technical reports, theses

To update publications:

Option 1: Automatic sync (recommended)

• Publications auto-sync weekly via GitHub Actions
• Or run manually: make sync-pubs / ./sync-pubs.sh

Option 2: Manual editing

1. Edit the appropriate .bib file directly
2. Run make build or latexmk -pdf sidd-cv.tex to rebuild
3. Verify changes in the generated PDF
4. Commit both .bib changes and sidd-cv.pdf

Note: The pubs/ directory is tracked directly in this repository (not a submodule). Publications are synced from personalrobotics/pubs.

Troubleshooting

"Missing CSV file" errors

If LaTeX compilation fails with missing CSV errors:

# Regenerate all CSV files from YAML
python3 yaml2csv.py --owner siddhss5 --repo siddhss5.github.io --branch main --output-dir data/

# Force rebuild
latexmk -pdf -g sidd-cv.tex

"PyYAML not found" error

pip install pyyaml

# If using system Python on macOS:
python3 -m pip install --user pyyaml

Bibliography not updating

If publication changes aren't reflected:

# Clean all auxiliary files
latexmk -c

# Remove bibliography cache
rm *.bbl *.blg *.aux

# Force full rebuild
latexmk -pdf -g sidd-cv.tex

GitHub Actions build failing

1. Check the Actions tab for detailed error logs
2. Verify YAML files are valid in the website repo
3. Ensure CV_REPO_PAT secret is configured in website repo (for cross-repo triggers)
4. Test locally with the same commands as CI uses

GitHub Actions

This repository includes automated PDF building via GitHub Actions:

Automatic Builds

• Triggers:
  • Push to main branch
  • Pull requests
  • Repository dispatch from website repo (when YAML files change)
  • Manual workflow dispatch
• Actions:
  • Fetches YAML files from website repo
  • Converts YAML to CSV using yaml2csv.py
  • Compiles PDF using latexmk
  • Commits updated PDF back to repo

Cross-Repository Automation

When YAML files are updated in the website repo:

1. Website repo workflow sends repository_dispatch event
2. CV repo workflow triggers automatically
3. Fresh CSV files are generated from latest YAML
4. PDF is rebuilt and committed

Workflow Files

• .github/workflows/build.yml - CV build workflow (this repo)
• .github/workflows/sync-pubs.yml - Weekly publication sync (this repo)
• .github/workflows/trigger-cv-build.yml - Trigger workflow (website repo)

Setup Requirements

The website repo needs a CV_REPO_PAT secret (Personal Access Token) with repo scope to trigger builds in this repo.

License

This CV template and data are for personal use by Siddhartha Srinivasa.