Shabnam Hakimi — CV in LaTeX (Awesome-CV)

Project Structure

cv/
├── cv.tex                      # Master file — all toggles and preamble patches live here
├── resume.tex                  # Shorter resume variant
├── awesome-cv.cls              # Awesome-CV class file
├── fonts/                      # SourceSans3 and Roboto TTF files (required by cls)
├── sections/
│   ├── education.tex
│   ├── experience.tex
│   ├── grants.tex
│   ├── talks.tex               # Uses cventries (not cvhonors) — see Talks section below
│   ├── teaching.tex
│   ├── service.tex
│   ├── reviewer.tex
│   ├── mentorship.tex
│   └── memberships.tex
├── refs/
│   ├── journals.bib            # @article entries (peer-reviewed journals)
│   ├── preprints.bib           # @unpublished entries
│   ├── conference.bib          # @inproceedings entries
│   ├── presentations.bib       # @misc entries (selected conference presentations)
│   ├── patents.bib             # @patent entries
│   └── scicomm.bib             # @misc entries (science communication)
└── scripts/
    ├── _shared.py              # Shared utilities (manual fingerprints, dedup logic)
    ├── update_refs.py          # Master orchestrator — runs all fetch scripts + dedup
    ├── fetch_orcid.py          # Fetch publications from ORCID
    ├── fetch_pubmed.py         # Fetch publications from PubMed
    ├── fetch_patents.py        # Fetch/discover patents from USPTO (interactive)
    ├── fetch_scholar.py        # Google Scholar fetch (optional; see note below)
    ├── add_ref.py             # Add a single reference manually (interactive)
    └── review_bib.py          # Review and edit entries in an existing .bib file

First-Time Setup

1. Install LaTeX

Requires TeX Live 2022+ or MiKTeX with XeLaTeX and the following packages: biblatex, biber, etoolbox, fontspec, fontawesome5, geometry, parskip

2. Fonts

The cls expects TTF files in a fonts/ directory alongside cv.tex:

File	Used for
`SourceSans3-Regular.ttf` (+ Bold, Italic, BoldItalic, Light, LightItalic, SemiBold, SemiBoldItalic)	Header / name
`Roboto-Regular.ttf` (+ Bold, Italic, BoldItalic, Thin, Light, Medium)	Body text

Download SourceSans3 from adobe-fonts/source-sans and Roboto from Google Fonts.

3. Install Python dependencies

make setup                    # installs requests, bibtexparser, pytest
pip install scholarly         # optional — for Google Scholar fetching

make check-deps verifies the required deps are importable before a run.

4. Configure scripts

Edit the CONFIG block at the top of each fetch script:

scripts/fetch_orcid.py

ORCID_ID = "0000-0003-4122-6041"

scripts/fetch_pubmed.py

PUBMED_AUTHOR = "Hakimi S"
ADMIN_EMAIL   = "you@example.com"   # required by NCBI policy

scripts/fetch_patents.py — requires a free USPTO Open Data Portal API key:

Register at developer.uspto.gov → "Sign Up"
After logging in, go to My Account → API Keys and generate a key
Add it to your shell profile so it persists across sessions:
```
echo 'export USPTO_API_KEY="your_key_here"' >> ~/.zshrc
source ~/.zshrc
```
The script reads os.environ["USPTO_API_KEY"] at runtime and exits with a clear error if it is not set.

Building the CV

One-command build:

make

⚠️ The Makefile references main.tex — update it to cv.tex if you haven't already.

Manual build sequence:

xelatex cv.tex
biber cv
xelatex cv.tex
xelatex cv.tex

Clean auxiliary files:

make clean
# or manually:
rm -f cv.aux cv.bbl cv.bcf cv.blg cv.log cv.out cv.run.xml cvbibcounts.tex

⚠️ Awesome-CV requires XeLaTeX, not pdflatex or lualatex.

Toggle Reference

All toggles are set near the top of cv.tex. Flip true/false to show or hide sections and control filtering.

Section visibility

Toggle	Default	Effect
`showJournals`	`true`	Peer-reviewed journal publications
`showPreprints`	`true`	Preprints & working papers
`showConferenceProceedings`	`true`	Conference proceedings
`showConferencePresentations`	`true`	Selected conference presentations
`showPatents`	`true`	Patents section
`showSciComm`	`true`	Science communication subsection
`showBookChapters`	`true`	Book chapters (if any)
`showTalks`	`true`	Invited talks subsection
`showGrants`	`true`	Grants & Awards section
`showTeaching`	`true`	Teaching section
`showService`	`true`	Professional Service section
`showReviewer`	`true`	Reviewing section
`showMentorship`	`true`	Mentorship section
`showMemberships`	`true`	Memberships section
`showReferences`	`true`	References section
`showFullReferences`	`false`	`true` = list referees; `false` = "Available upon request"
`showFullCVNote`	`false`	`true` = closing line "Full CV available upon request." (industry/speaking presets)

Selected-only filtering

When a selectedOnly toggle is true, only .bib entries with keywords = {selected} appear.

Toggle	Default	Effect
`selectedOnlyPublications`	`false`	Changes section heading to "Selected Publications"
`selectedOnlyJournals`	`false`	Filter journals to `keywords={selected}` only
`selectedOnlyPreprints`	`false`	Filter preprints
`selectedOnlyConferenceProceedings`	`false`	Filter conference proceedings
`selectedOnlyPresentations`	`true`	Filter presentations
`selectedOnlyPatents`	`false`	Filter patents (mirrors `\patentFilter`)

Show-all overrides

For sections that default to a curated subset:

Toggle	Default	Effect
`showAllTalks`	`false`	`true` = show all talks; `false` = "Selected Invited Talks"
`showAllGrants`	`false`	`true` = show all grants
`showAllTeaching`	`false`	`true` = show all teaching entries
`showAllService`	`false`	`true` = show all service entries

Bibliography numbering

Toggle	Default	Effect
`showBibNumbers`	`true`	Show [1], [2], … labels before each entry
`reverseBibNumbers`	`false`	`true` = N…1 (most recent first); `false` = 1…N

Patent filter

Patents use a separate \def rather than a toggle:

\def\patentFilter{selected}   % only keywords={selected} entries
\def\patentFilter{all}        % all entries

Marking Entries

Selected entries

Add keywords = {selected} to any .bib entry to include it in filtered views:

@article{Hakimi2021pairing,
  ...
  keywords = {selected},
}

Multiple keywords are comma-separated: keywords = {selected, manual}.

Manual / protected entries

Add keywords = {manual} (or include manual among other keywords) to prevent the deduplicator and fetch scripts from ever removing or overwriting that entry:

@article{Hakimi2015enhanced,
  ...
  keywords = {manual, selected},
}

Manual entries take priority over auto-fetched entries in all deduplication conflicts.

Keyword tagging during interactive review

Press [k] at any prompt during interactive review to add or edit keyword tags on the current entry before accepting or marking it. Tags are comma-separated; manual is always preserved when already present. Leave blank to keep existing tags, or enter - to clear all non-manual tags.

  Current keywords: (none)
  Add tags (comma-separated). Leave blank to keep current. Enter - to clear.
  Tags: selected, highlight

After tagging the entry is re-displayed so you can confirm before accepting.

Bolding Your Name in References

Name bolding is handled automatically by the boldifown name formatter defined in the cv.tex preamble. No manual .bib edits needed for standard entries.

For entries with complex or non-standard author lists (e.g. equal contributions, consortium papers), use the usera field to supply a pre-formatted author string:

@article{Hakimi2021pairing,
  author = {Hakimi, Shabnam and Sinclair, Adrienne H. and ...},
  usera  = {\cvbibname{Hakimi,~S}\cvfnmark{equalcontrib},
             Sinclair,~AH\cvfnmark{equalcontrib}, Stanley,~M, ...},
  ...
}

\cvbibname{...} — renders the name in bold (same weight as the auto-bold formatter)
\cvfnmark{equalcontrib} — renders a superscript *
When usera is present it replaces the normal \printnames{author} output entirely
When absent, \printnames{author} runs normally with boldifown applied

The * legend is printed once below the Publications section heading via \cvpublegend.

Talks Section

sections/talks.tex uses cventries / cventry (not cvhonors). This keeps the left margin aligned with all other sections.

Field mapping:

\cventry
  {Talk title in quotes}   % position field → appears on row 2, left
  {Venue name}             % title field    → appears on row 1, left (bold)
  {City, State}            % location field → appears on row 1, right (italic)
  {Year}                   % date field     → appears on row 2, right (italic)
  {}                       % description   → empty

Updating References

Fetch from all sources (ORCID + PubMed + Scholar) then dedup:

python scripts/update_refs.py
make fetch            # equivalent via Makefile

Dry run — preview without writing:

python scripts/update_refs.py --dry-run
make fetch-dry

Fetch from specific sources only:

python scripts/update_refs.py --sources orcid pubmed

Deduplicate only (no network fetch):

python scripts/update_refs.py --dedup-only
make dedup

Patents — manual step required:

The patent fetch script is interactive and must be run separately. Requires USPTO_API_KEY to be set in your environment (see First-Time Setup above).

python scripts/fetch_patents.py               # refresh known patents from USPTO
python scripts/fetch_patents.py --mode discover  # search for new patents by inventor name
python scripts/fetch_patents.py --dry-run     # preview changes

Reviewing Skipped & Rejected Entries

Before the interactive review loop, each fetch script silently filters entries in three categories. Two flags let you inspect and correct these filters.

What gets silently filtered

Filter	Matched by	Stored in
Already in `.bib`	DOI, PMID, or normalised title	The `.bib` file
Matches a `manual` fingerprint	cite key, DOI, PMID, or title	Entries with `keywords={manual}`
Previously rejected	DOI, PMID, or normalised title	`refs/.<source>_rejected.json`

`--show-skipped` — audit what was filtered and why

Prints a tagged line for every filtered entry during a normal fetch run:

[skip-exist]    10.1038/s41562-021-01234-5 already in refs/journals.bib
[skip-manual]   Hakimi2019reward matches a protected manual entry
[skip-rejected] 34127854: Neural correlates of adaptive learning...

Works per-source or across all sources:

python scripts/update_refs.py --show-skipped
python scripts/fetch_pubmed.py --show-skipped

`--review-rejected` — triage the rejection list interactively

Loads each source's refs/.<source>_rejected.json and presents entries one by one. For each you can keep it rejected or un-reject it (remove from the list). Un-rejected entries will reappear as candidates on the next normal fetch run.

# Review rejection lists for all sources
python scripts/update_refs.py --review-rejected

# Review a single source
python scripts/update_refs.py --review-rejected pubmed
python scripts/fetch_orcid.py --review-rejected
python scripts/fetch_scholar.py --review-rejected

# Preview without writing changes
python scripts/update_refs.py --review-rejected --dry-run

Tip: If you mis-rejected something, run --review-rejected to un-reject it, then run a normal fetch to pull it back in for interactive review.

Adding a Reference Manually

Use add_ref.py to add a single entry to any .bib file without running a full fetch. The entry is always written with at least keywords = {manual} so it is protected from deduplication.

Input modes

Mode	Flag	Description
DOI lookup	`--doi DOI`	Fetches metadata from Crossref, opens in `$EDITOR`
PMID lookup	`--pmid PMID`	Fetches metadata from PubMed, opens in `$EDITOR`
Manual entry	`--manual`	Opens a blank BibTeX template in `$EDITOR`
Fully interactive	(no flag)	Prompts for mode and `.bib` file interactively

Examples

# Fully interactive — prompts for mode, bib file, editor, and tags
python scripts/add_ref.py

# Pre-fill from Crossref by DOI
python scripts/add_ref.py --doi 10.1038/s41562-021-01234-5

# Pre-fill from PubMed by PMID
python scripts/add_ref.py --pmid 34127854

# Open a blank template; write to a specific bib file
python scripts/add_ref.py --manual --bib refs/conference.bib

# Preview without writing
python scripts/add_ref.py --doi 10.1038/s41562-021-01234-5 --dry-run

After editing in $EDITOR (fallback: nano), you are prompted for keyword tags. manual is always included automatically. Supported target .bib files: journals.bib, preprints.bib, conference.bib, presentations.bib, scicomm.bib, patents.bib.

Reviewing an Existing .bib File

Use review_bib.py to walk through entries already in a .bib file and edit, re-tag, or delete them interactively. This is separate from the fetch pipeline — it operates purely on what is already on disk.

Actions

Key	Action
`e`	Open entry in `$EDITOR`; re-display after saving
`k`	Add/edit keyword tags interactively; re-display
`m`	Toggle `manual` on/off in keywords; re-display
`d`	Mark for deletion (confirmed as a batch before writing)
`s`	Skip — leave unchanged, move to next entry
`o`	Open DOI in browser (shown only when entry has a DOI field)
`q`	Stop reviewing; write all changes made so far

e, k, and m re-display the entry after acting so you can keep editing before moving on. Deletions are collected and shown as a confirmation summary before anything is written. Every write is preceded by a timestamped .bak backup of the original file.

Examples

# Review all entries in a bib file
python scripts/review_bib.py refs/journals.bib

# Review only entries tagged 'selected'
python scripts/review_bib.py refs/conference.bib --filter selected

# Resume after quitting at entry 12
python scripts/review_bib.py refs/journals.bib --start 12

# Preview changes without writing
python scripts/review_bib.py refs/preprints.bib --dry-run

All six .bib files are supported: journals.bib, preprints.bib, conference.bib, presentations.bib, scicomm.bib, patents.bib.

Google Scholar

Google Scholar blocks automated scraping. fetch_scholar.py will:

Skip silently with a helpful message if scholarly is not installed (exit 0)
Attempt a fetch if scholarly is installed
Warn and exit cleanly (exit 0) if Scholar rate-limits or blocks the request, so it never causes update_refs.py to report a hard error

To reduce rate-limit risk, set a ScraperAPI key in fetch_scholar.py:

SCRAPER_API_KEY = "your_key_here"

For a reliable manual alternative:

Go to your Google Scholar profile
Select all publications → Export → BibTeX
Append to the appropriate .bib file and add keywords = {manual}
Run python scripts/update_refs.py --dedup-only to remove any duplicates

How `_shared.py` Works

scripts/_shared.py contains utilities shared across all fetch scripts:

load_manual_fingerprints(bib_path) — reads a .bib file and returns a dict of {keys, dois, pmids, titles} for all keywords={manual} entries
fingerprint_matches(entry, fp) — returns True if a candidate fetched entry matches any manual fingerprint (by cite key, DOI, PMID, or title)
is_manual(entry) — returns True if an entry's keywords contain "manual"
interactive_review(candidates, rejected_file) — walks through candidate BibTeX entries with actions a/e/k/m/r/s/o (accept, edit, keywords, manual, reject+remember, skip, open DOI)
review_rejected(rejected_file) — interactively triages a rejection list; un-rejected entries are removed from the JSON so they reappear on the next fetch
prompt_keywords(current) / inject_keywords(bibtex, kw) — interactive keyword tagging helpers used by both the review loop and add_ref.py

Each fetch script imports these at startup:

sys.path.insert(0, str(Path(__file__).parent))
from _shared import load_manual_fingerprints, fingerprint_matches

This ensures the functions are available whether scripts are run standalone (python scripts/fetch_orcid.py) or as subprocesses via update_refs.py.

Color Themes

Change \colorlet{awesome}{...} in cv.tex to one of: awesome-emerald, awesome-skyblue, awesome-red, awesome-pink, awesome-orange, awesome-nephritis, awesome-concrete, awesome-darknight

Or use a custom color:

\colorlet{awesome}{black}    % current setting — monochrome

Name		Name	Last commit message	Last commit date
.github/workflows		.github/workflows
config		config
fonts		fonts
refs		refs
scripts		scripts
sections		sections
tests		tests
.gitignore		.gitignore
CLAUDE.md		CLAUDE.md
README.md		README.md
awesome-cv-resume.cls		awesome-cv-resume.cls
awesome-cv.cls		awesome-cv.cls
coverletter.tex		coverletter.tex
cv-preset.tex		cv-preset.tex
cv.tex		cv.tex
makefile		makefile
requirements-dev.txt		requirements-dev.txt
requirements.txt		requirements.txt
resume.tex		resume.tex

Latest commit

History

Folders and files

Repository files navigation

Shabnam Hakimi — CV in LaTeX (Awesome-CV)

Project Structure

First-Time Setup

1. Install LaTeX

2. Fonts

3. Install Python dependencies

4. Configure scripts

Building the CV

One-command build:

Manual build sequence:

Clean auxiliary files:

Toggle Reference

Section visibility

Selected-only filtering

Show-all overrides

Bibliography numbering

Patent filter

Marking Entries

Selected entries

Manual / protected entries

Keyword tagging during interactive review

Bolding Your Name in References

Talks Section

Updating References

Fetch from all sources (ORCID + PubMed + Scholar) then dedup:

Dry run — preview without writing:

Fetch from specific sources only:

Deduplicate only (no network fetch):

Patents — manual step required:

Reviewing Skipped & Rejected Entries

What gets silently filtered

--show-skipped — audit what was filtered and why

--review-rejected — triage the rejection list interactively

Adding a Reference Manually

Input modes

Examples

Reviewing an Existing .bib File

Actions

Examples

Google Scholar

How _shared.py Works

Color Themes

About

Resources

Stars

Watchers

Forks

Releases

Packages

Contributors

Languages

`--show-skipped` — audit what was filtered and why

`--review-rejected` — triage the rejection list interactively

How `_shared.py` Works