In this demo I cover the fundamental end-to-end workflow involved in
creating and managing tailored job applications using
autocv.
A minimal reproducible example
The following end-to-end example runs in a temporary environment on
your local machine, constructing a project directory, an application
directory, a resume, and keyword reports, all from the
autocv-supplied example job posting, template datasets, and
template stylesheets.
autocv::in_tmp_env({
message("1. Build the project directory...")
autocv::build_base_directory()
data(example_job_metadata)
app_df <- autocv::construct_app_metadata(app_info = example_job_metadata)
message("")
message("2. Build the application...")
autocv::build_app_directory(app_df = app_df, open = FALSE)
message("")
message("3. Render the application...")
# autocv::render_app(cover = FALSE, email = FALSE)
autocv::render_resume_plain()
message("")
message("4. Check application keywords...")
report_df <- autocv::check_skills()
print(report_df)
message("")
message("5. Inspect application metadata...")
app_df <- autocv::get_app_info()
print(app_df)
})A step-by-step example
Let us now walk through each of these steps in greater depth, keeping track of the evolving file tree and copying the generated documents from the temporary environment to your working directory for reference later.
Step 1: Build the project directory
We begin by creating the basic file tree upon which to build tailored
job applications. The file tree includes placeholder
output/ and applications/ directories that
store generated documents, as well as an input/ directory
containing the template spreadsheets resume_data.xlsx
(see example_skill_data
and example_position_data),
cover_data.xlsx
(see example_contact_data
and example_text_data),
and job_metadata.yml (example_job_metadata).
A .Rprofile file containing path environment variables is
automatically generated, along with an Rproj file and the
notebooks resume.Rmd
and cover.Rmd,
which are used to render job application documents from data contained
in the aforementioned spreadsheets.
autocv::in_tmp_env({
autocv::build_base_directory()
})
#>
#> ── Setting project paths ───────────────────────────────────────────────────────
#>
#> Your current path is: /tmp/RtmpkTOdSk/
#> Set your desired project path relative to your current path:
#>
#> You entered the path: /tmp/RtmpkTOdSk/
#>
#> Setting your project root to: /tmp/RtmpkTOdSk/
#> Created file .here in /tmp/RtmpkTOdSk . Please start a new R session in the new project directory.
#> ✔ Writing file: .Rprofile
#> ✔ Writing file: mycv.Rproj
#>
#> ℹ Writing to: ../../../../../../tmp/RtmpkTOdSk
#>
#> ── Creating directory tree ─────────────────────────────────────────────────────
#> ! Folder already exists: . (skipping)
#> ✔ Creating folder: R
#> ✔ Creating folder: R/input
#> ✔ Creating folder: R/output
#> ✔ Creating folder: R/applications
#>
#> ── Writing skeleton data files ─────────────────────────────────────────────────
#> ✔ Writing file: R/input/resume_data.xlsx
#> ✔ Writing file: R/input/cover_data.xlsx
#> ✔ Writing file: R/input/job_metadata.yml
#>
#> ── Writing resume-building notebooks ───────────────────────────────────────────
#> ✔ Writing file: resume.Rmd
#> ✔ Writing file: cv.Rmd#> Current file tree:
#> /tmp/RtmpkTOdSk/.
#> ├── R
#> │ ├── applications
#> │ ├── input
#> │ │ ├── cover_data.xlsx
#> │ │ ├── job_metadata.yml
#> │ │ └── resume_data.xlsx
#> │ └── output
#> ├── cv.Rmd
#> ├── filea44b717a448b
#> ├── filea44b7c562b13
#> ├── filea44bba5b49
#> ├── filea44bc33638
#> ├── libloc_177_2e46c0fc1ffb1091.rds
#> ├── libloc_239_283e8cdc8a036cdc.rds
#> ├── libloc_246_4744b1836fb1a8fb.rds
#> ├── mycv.Rproj
#> ├── resume.Rmd
#> └── rmarkdown-stra44b54b49e21.htmlStep 2: Build the application directory
With the project directory constructed and template data files copied
over, we can now generate tailored job applications. To do so, you
should update the fields in the input/job_metadata.yml file
with information relevant to the job of interest, namely the
id, company, position, and
posting_url fields, as well as optionally populating the
base_id, portal_url,
linkedin_url, recruiter_email, and
notes fields. For additional information, see example_job_metadata.
autocv::in_tmp_env({
data(example_job_metadata)
app_df <- autocv::construct_app_metadata(app_info = example_job_metadata)
autocv::build_app_directory(app_df = app_df, open = FALSE)
})
#>
#> ℹ Writing to: ../../../../../../tmp/RtmpkTOdSk/R/applications/2024-07-data-science/2024-09-28-01-company-name-data-scientist-AB
#>
#> ── Creating directory tree ─────────────────────────────────────────────────────
#> ✔ Creating folder: input
#> ✔ Creating folder: output
#>
#> ── Copying base data files into directory ──────────────────────────────────────
#> ✔ Writing file: input/resume_data_AB.xlsx
#> ✔ Writing file: input/cover_data_AB.xlsx
#>
#> ── Writing application metadata ────────────────────────────────────────────────
#> ✔ Writing file: metadata_AB.yml
#> ✔ Writing file: ../log.rds
#>
#> ── Downloading job posting and building a skill report ─────────────────────────
#> ✔ Writing file: input/posting_AB.txt
#>
#> ── Keyword check: Posting vs job terms list ────────────────────────────────────
#> ✔ Writing file: output/keyword_counts_AB.csv
#> # A tibble: 1 × 2
#> term matches
#> <chr> <dbl>
#> 1 Disabled 4
#>
#> ── Keyword check: Posting vs data terms list ───────────────────────────────────
#> ✔ Writing file: output/skill_counts_posting_AB.csv
#> # A tibble: 55 × 2
#> term matches
#> <chr> <dbl>
#> 1 Data Science 73
#> 2 Data Analysis 15
#> 3 UX 1
#> 4 Statistics 29
#> 5 Cloud Computing 4
#> 6 Insights 10
#> 7 Computer Science 7
#> 8 Survey Data 1
#> 9 Algorithms 4
#> 10 Unstructured Data 2
#> 11 Domain Knowledge 3
#> 12 Mathematics 3
#> 13 Data-Driven 3
#> 14 Statistical Knowledge 1
#> 15 Data Visualization 3
#> 16 Machine Learning 8
#> 17 C 8
#> 18 Classification 3
#> 19 Data Mining 4
#> 20 Statistical Learning 3
#> 21 Data Collection 1
#> 22 Big Data 13
#> 23 Exploratory Data Analysis 2
#> 24 Data Management 2
#> 25 Data Cleaning 1
#> 26 Predictive Models 3
#> 27 Models 2
#> 28 Data-Driven Decisions 1
#> 29 Statistical Analysis 2
#> 30 Data Preprocessing 1
#> 31 Feature Engineering 1
#> 32 Model Selection 1
#> 33 Implementation 1
#> 34 Analytical Techniques 1
#> 35 Analytical 1
#> 36 Large Datasets 2
#> 37 Non-Technical Audiences 1
#> 38 Critical Thinking 1
#> 39 Data-Driven Decision-Making 1
#> 40 Data Analytics 1
#> 41 Cloud Services 1
#> 42 Python 1
#> 43 R 2
#> 44 Data Engineering 1
#> 45 HTML 4
#> 46 Statistical Modeling 1
#> 47 Causal Inference 1
#> 48 Collaboration 1
#> 49 Signal Processing 1
#> 50 Spark 1
#> 51 Spark SQL 1
#> 52 Data Ethics 1
#> 53 ETL 1
#> 54 ELT 1
#> 55 Developers 1
#>
#> ✔ Opening: .#> Current file tree:
#> /tmp/RtmpkTOdSk/./R/applications
#> └── 2024-07-data-science
#> ├── 2024-09-28-01-company-name-data-scientist-AB
#> │ ├── input
#> │ │ ├── cover_data_AB.xlsx
#> │ │ ├── posting_AB.txt
#> │ │ ├── report.xlsx
#> │ │ └── resume_data_AB.xlsx
#> │ ├── metadata_AB.yml
#> │ └── output
#> │ ├── keyword_counts_AB.csv
#> │ └── skill_counts_posting_AB.csv
#> └── log.rdsStep 3: Edit the application
After building the application directory, run
autocv::edit_app(). This will open the spreadsheets, namely
resume_data.xlsx
and cover_data.xlsx,
containing the data that will be used to generate your application
documents. Edit the entries and metadata in these documents as desired
to suit the job in question.
Step 4: Render the application
Once you’ve tailored the information in your application to the job in question, you are ready to render your application documents.
autocv::in_tmp_env({
# autocv::render_app(cover = FALSE, email = FALSE)
autocv::render_resume_plain()
})
#>
#> ── Building resume.txt ─────────────────────────────────────────────────────────
#> ✔ Writing file: output/resume_yourname_AB.txt#> Current file tree:
#> /tmp/RtmpkTOdSk/./R/applications
#> └── 2024-07-data-science
#> ├── 2024-09-28-01-company-name-data-scientist-AB
#> │ ├── input
#> │ │ ├── cover_data_AB.xlsx
#> │ │ ├── posting_AB.txt
#> │ │ ├── report.xlsx
#> │ │ └── resume_data_AB.xlsx
#> │ ├── metadata_AB.yml
#> │ └── output
#> │ ├── keyword_counts_AB.csv
#> │ ├── resume_yourname_AB.txt
#> │ └── skill_counts_posting_AB.csv
#> └── log.rdsStep 5: Check application keywords
Having generated a resume, we can now analyze its contents to determine which data-related keywords we included and whether they align with the keywords in the job posting. This step performs three independent checks – identify general job-related keywords in the job posting, identify data science keywords in the job posting, identify data science keywords in the resume – followed by a keyword match of the data science terms from the job posting against those mentioned in the resume, either in the skills section or in the body text.
autocv::in_tmp_env({
report_df <- autocv::check_skills()
print(report_df)
})
#>
#> ── Keyword check: Posting vs job terms list ────────────────────────────────────
#> ! File already exists: output/keyword_counts_AB.csv (skipping)
#> term matches
#> 1 Disabled 4
#>
#> ── Keyword check: Posting vs data terms list ───────────────────────────────────
#> ! File already exists: output/skill_counts_posting_AB.csv (skipping)
#> term matches
#> 1 Data Science 73
#> 2 Data Analysis 15
#> 3 UX 1
#> 4 Statistics 29
#> 5 Cloud Computing 4
#> 6 Insights 10
#> 7 Computer Science 7
#> 8 Survey Data 1
#> 9 Algorithms 4
#> 10 Unstructured Data 2
#> 11 Domain Knowledge 3
#> 12 Mathematics 3
#> 13 Data-Driven 3
#> 14 Statistical Knowledge 1
#> 15 Data Visualization 3
#> 16 Machine Learning 8
#> 17 C 8
#> 18 Classification 3
#> 19 Data Mining 4
#> 20 Statistical Learning 3
#> 21 Data Collection 1
#> 22 Big Data 13
#> 23 Exploratory Data Analysis 2
#> 24 Data Management 2
#> 25 Data Cleaning 1
#> 26 Predictive Models 3
#> 27 Models 2
#> 28 Data-Driven Decisions 1
#> 29 Statistical Analysis 2
#> 30 Data Preprocessing 1
#> 31 Feature Engineering 1
#> 32 Model Selection 1
#> 33 Implementation 1
#> 34 Analytical Techniques 1
#> 35 Analytical 1
#> 36 Large Datasets 2
#> 37 Non-Technical Audiences 1
#> 38 Critical Thinking 1
#> 39 Data-Driven Decision-Making 1
#> 40 Data Analytics 1
#> 41 Cloud Services 1
#> 42 Python 1
#> 43 R 2
#> 44 Data Engineering 1
#> 45 HTML 4
#> 46 Statistical Modeling 1
#> 47 Causal Inference 1
#> 48 Collaboration 1
#> 49 Signal Processing 1
#> 50 Spark 1
#> 51 Spark SQL 1
#> 52 Data Ethics 1
#> 53 ETL 1
#> 54 ELT 1
#> 55 Developers 1
#>
#> ── Keyword check: Resume vs data terms list ────────────────────────────────────
#> ✔ Writing file: output/skill_counts_resume_AB.csv
#> # A tibble: 16 × 2
#> term matches
#> <chr> <dbl>
#> 1 GitHub 3
#> 2 Python 2
#> 3 R 2
#> 4 SQL 2
#> 5 Machine Learning 1
#> 6 SciPy 3
#> 7 Scikit-Learn 2
#> 8 TensorFlow 2
#> 9 DevOps 1
#> 10 Git 1
#> 11 GitHub Actions 2
#> 12 ML Pipelines 1
#> 13 Data Modeling 1
#> 14 CI/CD Pipelines 1
#> 15 Bayesian Inference 1
#> 16 Documentation 2
#>
#> ── Keyword report: Resume vs posting keywords ──────────────────────────────────
#> Not in your skill set:
#> 1. Algorithms
#>
#> 2. Analytical
#>
#> 3. Analytical Techniques
#>
#> 4. C
#>
#> 5. Cloud Computing
#>
#> 6. Cloud Services
#>
#> 7. Collaboration
#>
#> 8. Computer Science
#>
#> 9. Data Analysis
#>
#> 10. Data Analytics
#>
#> 11. Data Cleaning
#>
#> 12. Data Collection
#>
#> 13. Data Engineering
#>
#> 14. Data Ethics
#>
#> 15. Data Mining
#>
#> 16. Data Science
#>
#> 17. Data-Driven
#>
#> 18. Data-Driven Decision-Making
#>
#> 19. Data-Driven Decisions
#>
#> 20. Developers
#>
#> 21. Domain Knowledge
#>
#> 22. ELT
#>
#> 23. Implementation
#>
#> 24. Insights
#>
#> 25. Large Datasets
#>
#> 26. Mathematics
#>
#> 27. Models
#>
#> 28. Non-Technical Audiences
#>
#> 29. Predictive Models
#>
#> 30. Signal Processing
#>
#> 31. Spark SQL
#>
#> 32. Statistical Knowledge
#>
#> 33. Survey Data
#>
#> 34. Unstructured Data
#>
#> 35. UX
#>
#>
#>
#> ✔ Writing file: output/skill_report_AB.csv
#>
#>
#>
#> • 5.45% of the posting keywords are in your resume.
#> term count matches in_my_skill_set in_my_skill_list
#> 1 Machine Learning 8 1 TRUE FALSE
#> 2 Python 1 2 TRUE TRUE
#> 3 R 2 2 TRUE TRUE
#> 4 Data Science 73 0 FALSE FALSE
#> 5 Data Analysis 15 0 FALSE FALSE
#> 6 UX 1 0 FALSE FALSE
#> 7 Statistics 29 0 TRUE FALSE
#> 8 Cloud Computing 4 0 FALSE FALSE
#> 9 Insights 10 0 FALSE FALSE
#> 10 Computer Science 7 0 FALSE FALSE
#> 11 Survey Data 1 0 FALSE FALSE
#> 12 Algorithms 4 0 FALSE FALSE
#> 13 Unstructured Data 2 0 FALSE FALSE
#> 14 Domain Knowledge 3 0 FALSE FALSE
#> 15 Mathematics 3 0 FALSE FALSE
#> 16 Data-Driven 3 0 FALSE FALSE
#> 17 Statistical Knowledge 1 0 FALSE FALSE
#> 18 Data Visualization 3 0 TRUE FALSE
#> 19 C 8 0 FALSE FALSE
#> 20 Classification 3 0 TRUE FALSE
#> 21 Data Mining 4 0 FALSE FALSE
#> 22 Statistical Learning 3 0 TRUE FALSE
#> 23 Data Collection 1 0 FALSE FALSE
#> 24 Big Data 13 0 TRUE FALSE
#> 25 Exploratory Data Analysis 2 0 TRUE FALSE
#> 26 Data Management 2 0 TRUE FALSE
#> 27 Data Cleaning 1 0 FALSE FALSE
#> 28 Predictive Models 3 0 FALSE FALSE
#> 29 Models 2 0 FALSE FALSE
#> 30 Data-Driven Decisions 1 0 FALSE FALSE
#> 31 Statistical Analysis 2 0 TRUE FALSE
#> 32 Data Preprocessing 1 0 TRUE FALSE
#> 33 Feature Engineering 1 0 TRUE FALSE
#> 34 Model Selection 1 0 TRUE FALSE
#> 35 Implementation 1 0 FALSE FALSE
#> 36 Analytical Techniques 1 0 FALSE FALSE
#> 37 Analytical 1 0 FALSE FALSE
#> 38 Large Datasets 2 0 FALSE FALSE
#> 39 Non-Technical Audiences 1 0 FALSE FALSE
#> 40 Critical Thinking 1 0 TRUE FALSE
#> 41 Data-Driven Decision-Making 1 0 FALSE FALSE
#> 42 Data Analytics 1 0 FALSE FALSE
#> 43 Cloud Services 1 0 FALSE FALSE
#> 44 Data Engineering 1 0 FALSE FALSE
#> 45 HTML 4 0 TRUE FALSE
#> 46 Statistical Modeling 1 0 TRUE FALSE
#> 47 Causal Inference 1 0 TRUE FALSE
#> 48 Collaboration 1 0 FALSE FALSE
#> 49 Signal Processing 1 0 FALSE FALSE
#> 50 Spark 1 0 TRUE FALSE
#> 51 Spark SQL 1 0 FALSE FALSE
#> 52 Data Ethics 1 0 FALSE FALSE
#> 53 ETL 1 0 TRUE FALSE
#> 54 ELT 1 0 FALSE FALSE
#> 55 Developers 1 0 FALSE FALSE#> Current file tree:
#> /tmp/RtmpkTOdSk/./R/applications
#> └── 2024-07-data-science
#> ├── 2024-09-28-01-company-name-data-scientist-AB
#> │ ├── input
#> │ │ ├── cover_data_AB.xlsx
#> │ │ ├── posting_AB.txt
#> │ │ ├── report.xlsx
#> │ │ └── resume_data_AB.xlsx
#> │ ├── metadata_AB.yml
#> │ └── output
#> │ ├── keyword_counts_AB.csv
#> │ ├── resume_yourname_AB.txt
#> │ ├── skill_counts_posting_AB.csv
#> │ ├── skill_counts_resume_AB.csv
#> │ └── skill_report_AB.csv
#> └── log.rdsThe following spreadsheets were generated by this step:
-
keyword_counts_AB.csv: General keyword counts for the job posting -
skill_counts_posting_AB.csv: Data-related keyword counts for the job posting -
skill_counts_resume_AB.csv: Data-related keyword counts for the generated resume -
skill_report_AB.csv: Job posting keywords included in the generated resume
Step 6: Inspect application metadata
Every step so far has yielded metadata that was automatically stored in a log file, with each row corresponding to a unique job application. If at any point you wish to review the information for an existing job application, you can easily retrieve it from the log.
autocv::in_tmp_env({
app_df <- autocv::get_app_info()
print(app_df)
})
#> id company position status date_applied days_since notes
#> 1 AB Company Name Data Scientist ipr / / /Step 7: Update application metadata
At this point we can update the job application log with new information as desired, such as changes to the status of the application or a note about the job in question.
autocv::in_tmp_env({
autocv::apply_to(app_id = "AB")
autocv::update_app_info(app_id = "AB", notes = "Excited about this role")
app_df <- autocv::get_app_info()
print(app_df)
})
#> • Modifying field: status (ipr -> applied)
#> • Modifying field: date_applied (/ -> 2024-09-28)
#>
#> ! File already exists: metadata_AB.yml (overwriting)
#> ! File already exists: ../log.rds (updating entry for 'AB')
#> • Modifying field: notes (/ -> Excited about this role)
#>
#> ! File already exists: metadata_AB.yml (overwriting)
#> ! File already exists: ../log.rds (updating entry for 'AB')
#> id company position status date_applied days_since
#> 1 AB Company Name Data Scientist applied 2024-09-28 0
#> notes
#> 1 Excited about thi...Review the generated documents
Let’s conclude by transferring some of documents generated within our
temporary environment into long-term memory. Each
autocv-generated application document is displayed below
for reference.
autocv::in_tmp_env({
source_paths <- as.character(autocv::get_app_info(
id = "AB",
field = c(
"resume_plain_path",
"keyword_counts_posting",
"skill_counts_posting",
"skill_counts_resume",
"skill_report"
)
)[2:6])
target_dir <- file.path(fs::path_wd(), "output")
if (!dir.exists(target_dir)) { dir.create(target_dir) }
target_paths <- file.path(target_dir, basename(source_paths))
autocv::copy_files(
source_paths = source_paths,
target_paths = target_paths
)
})
#> ! File already exists: output/resume_yourname_AB.txt (skipping)
#> ! File already exists: output/keyword_counts_AB.csv (skipping)
#> ! File already exists: output/skill_counts_posting_AB.csv (skipping)
#> ! File already exists: output/skill_counts_resume_AB.csv (skipping)
#> ! File already exists: output/skill_report_AB.csv (skipping)#> Working directory output file tree:
#> /home/runner/work/autocv/autocv/vignettes/output
#> ├── cv_yourname.html
#> ├── keyword_counts_AB.csv
#> ├── resume_yourname_AB.pdf
#> ├── resume_yourname_AB.txt
#> ├── skill_counts_posting_AB.csv
#> ├── skill_counts_resume_AB.csv
#> └── skill_report_AB.csv
# NOTE: eval is off due to LaTeX rendering issues when running GitHub workflow
autocv::in_tmp_env({
autocv::render_cv_as_html()
autocv::render_resume()
})
autocv::in_tmp_env({
source_paths <- c(
as.character(autocv::get_app_info(id = "AB", field = "resume_path")[2]),
file.path(autocv::get_path_to("output"), "cv_yourname.html")
)
target_dir <- file.path(fs::path_wd(), "output")
if (!dir.exists(target_dir)) { dir.create(target_dir) }
target_paths <- file.path(target_dir, basename(source_paths))
autocv::copy_files(
source_paths = source_paths,
target_paths = target_paths
)
})
filename <- "cv_yourname.html"
source_path <- file.path(
autocv::get_path_to("notebooks"), "vignettes", "output", filename
)
target_path <- file.path(
autocv::get_path_to("notebooks"), "docs", filename
)
file.copy(source_path, target_path, overwrite = TRUE)
#> [1] TRUEThe following documents were generated by autocv:
-
resume_yourname_AB.pdf(PDF resume):
-
cv_yourname.html(web-based CV):
-
resume_yourname_AB.txt(plain text resume):
And there you have it! A few steps (“build, edit, render, check” and
“inspect, update”), neatly encapsulated within autocv’s
core user interface, are all you need to construct beautiful, tailored,
easily tracked, resume documents and job applications.