Coding DojoFive years of coding bootcamps in Albania, cohort after cohort.

Evidence from past work, not proof of the current product

This case applies to the engagement and period it describes. It does not prove outcomes of the microcourses, training, workshops, qualification or human evaluation that betaPlan offers today.

Cohorts run
34
Registered learners
~460
Stacks taught
5
Certified
~400

Over the years

The same path, run 34 times.

We taught the Coding Dojo curriculum in Albania from May 2021. Each cohort followed a four-month programme with three sessions a week, one stack from start to finish, and a graded certification exam at the end. As a licensed training centre, we reported every cohort to the National Agency for Employment and Skills: registered learners on one form, certified learners on another. This page sets out what those forms contain.

The first cohort started on 24 May 2021 with 5 learners in Java. 2022 brought 7 cohorts across Python, Java, C# .NET and JavaScript. 2023 was the peak, 15 cohorts and 176 registered learners in a single year, in Tiranë, Korçë and Shkodër. 2024 added 7 cohorts and 2025 closed the record with 2, the last one certified in July 2025.

Every cohort left the same two records. A registration form with the number of learners, how many were women, how many were registered as unemployed, their age group and their education, and a certification form listing every learner who finished and passed. About one in three registered learners was a woman. Every figure on this page comes from those forms.

Registered to certified

From registration to a certificate.

Each stack ends with a certification exam graded out of 10. A learner who finishes the four months and passes the exam is certified and reported by name to the agency. Names stay in the forms; this page gives only the counts.

  1. Registered

    ~460 (100%)

    May 2021 to July 2025

  2. Certified

    ~400 (87%)

    Nearly 9 in 10 of those registered

Year by year

Cohorts, registrations and certificates per year.

Counts from the quarterly forms. Cohorts and registrations are counted in the year the cohort started; certificates are counted in the year they were reported, which for an autumn cohort is the following year. This table reports the cohorts covered by the quarterly agency filing; the rounded totals above also include cohorts we ran outside that filing, all of them certified.

YearCohorts startedRegisteredCertified
202131917
202279544
202315176117
2024779104
202522552

The year-by-year table above comes to about 394 registered and 332 to 334 certified. The headline totals of 460 and 400 also include cohorts we ran outside the quarterly agency filing, all of them certified.

The outcome

What the records show.

Across five years and five stacks, Python, JavaScript, Java, C# .NET and Data Science, about 400 learners were certified. Python and JavaScript account for more than two thirds of them. Over the whole period, the certificates in the table come to 84 to 85% of its registrations, 332 to 334 of 394. Because certificates are counted in the year they were reported, the ratio of a single year does not describe a cohort: 2022 and 2023 show fewer certificates than registrations, 2024 and 2025 show more, as the autumn cohorts of the previous year were certified.

Alumni of the 2021 cohorts were followed after graduation. Of the 35 alumni the follow-up reached, 30 had graduated and 25 were recorded as employed at the follow-up, 21 of them in the month they graduated. The follow-up is a separate record from the agency forms and is not limited to the cohorts in the quarterly filing, which is why it reaches more people than the 2021 row of the table. That is a record of one year, not a promise about the next.

Courses at this scale, documented.

Each of these past cohorts left registration, attendance and certification records filed with a public agency. This shows only the discipline with which we ran courses in the past; it does not confer accreditation or prove outcomes for a current betaPlan offer.