Online Python and AI classes in Ayacucho, Peru

INEI: table 1.9 of Perú: Perfil Sociodemográfico 2017 for the city; table 1 of the Censos Nacionales 2025 demographic indicators for the rest. No 2025 city figure had been published as of November 2026. The units overlap and are kept apart.

Figures from tables 11.13, 11.6 and 11.5 of the first volume of INEI's 2026 statistical compendium, based on the ENAHO survey. Decimal commas changed to points. The 2015 comparison is ours.

Three in four people over six in the department now use the internet, up from about one in four in 2015. But only about one home in three has its own internet service and one in five has a computer. Many Ayacucho learners go online by phone or at a university lab, which shapes our advice on equipment further down.

Path, courses and outcomes are ours. The OPI, UTEC's free computing olympiad for secundaria students, also accepts Python in its first round, which gives teenagers one more reason to start with it.

Engineering students will find more on the subject in Python for engineering students in Peru, and in Spanish in Python para estudiantes de ingeniería.

1. A computer with a keyboard. Any laptop or desktop from the last several years is enough. Python is free and light. Typing code on a phone is not practical, and we will say so.
2. A connection that holds a video call. Speed matters less than steadiness. If home internet is not available, a university lab, a relative's home or a quiet internet café hour can work.
3. A headset with a microphone. It makes the mentor clearer and lets you speak without the whole house hearing.
4. A regular hour. Because we teach at any time you book, choose an hour when the computer and the connection are reliably free.
5. Patience with English. Mentors slow down and rephrase. You do not need fluent English; you need to be willing to listen and ask.

INEI's 2025 figures put a computer in about one Ayacucho home in five. If you do not have one, tell us at the free class. We would rather plan around a shared or borrowed machine, or tell you a course will not work yet, than sell a place you cannot use.

Why we insist on reasoning rather than copied answers, in any language, is set out in learn to think, not copy.

import csv

with open("matricula-2024-1-UNSCH.csv", encoding="utf-8") as data:
    students = list(csv.DictReader(data, delimiter=";"))
print(len(students))

Byte 0xcd is how Latin-1 stores the letter Í, as in INGENIERÍA. In UTF-8 that same byte means "a two-byte letter starts here", and the next byte does not fit, so Python refuses. Learners often search for a quick way past the error and find errors="replace". It makes the error go away, which is exactly the danger.

import csv

with open("matricula-2024-1-UNSCH.csv", encoding="utf-8", errors="replace") as data:
    students = list(csv.DictReader(data, delimiter=";"))
print(len(students))
print(students[0]["ESCUELA_PROFESIONAL"])
print("INGENIERÍA CIVIL" in [s["ESCUELA_PROFESIONAL"] for s in students])

The program now runs and counts every student, but each Í and Ó has become the replacement mark �. Worse, a search for the Civil Engineering school answers False, as if no one studied it. Nothing crashed, and the answer is wrong. That is the lesson in one line: a program that runs is not a program that is right.

import csv
from collections import Counter

with open("matricula-2024-1-UNSCH.csv", encoding="latin-1") as data:
    students = list(csv.DictReader(data, delimiter=";"))

schools = Counter(s["ESCUELA_PROFESIONAL"] for s in students)
print("Students:", len(students), "in", len(schools), "schools")
for name, n in schools.most_common(3):
    print(n, name)
print("INGENIERÍA CIVIL" in schools)

accented = Counter(ch for s in students for ch in s["ESCUELA_PROFESIONAL"] if ord(ch) > 127)
print("Letters outside plain ASCII:", dict(accented))

Data: Universidad Nacional de San Cristóbal de Huamanga, Estudiantes matriculados, on datosabiertos.gob.pe, Open Data Commons Attribution Licence (ODC-By). Counts are our own, from 11,993 rows across 28 professional schools. By birthplace, 11,620 students were born in the Ayacucho department, 204 in Cusco and 71 in Apurímac. Students' dates of birth, grades and other personal columns are never read.

With encoding="latin-1" every letter comes back and Civil Engineering is found. The last line counts the letters that needed care: 8,487 Í, 2,243 Ó and 108 Á in the school names alone. One more detail rewards a sharp eye: the faculty "INGENIERIA QUÍMICA Y METALURGIA" is typed without the accent on its first word, while the faculty next to it has one. So even with the right encoding, a search for "INGENIERÍA" in the faculty column misses 707 students. Real data needs checking twice.

1. Ages 9 to 12. Type the words Perú, año and Huamanga in a text editor, save the file in two different encodings, and open each one as if it were the other. Which letters survive?
2. Ages 13 to 18. Run Version 1, read the error aloud, and explain "byte 0xcd" in your own words before running Version 2.
3. University students. Count students by faculty. Then write a check that warns when two faculty names differ only by an accent.
4. Adults. Write a function that tries UTF-8 first and Latin-1 second, and reports which one worked, so your scripts never guess silently.

Peru stays on UTC minus 5 hours all year, without summer time since 1994 (tz database zone America/Lima; INACAL keeps official time). India runs five and a half hours ahead of UTC, so it is always 10 hours 30 minutes ahead of Ayacucho.

With neither clock changing, a booked hour stays the same through the year. Learners book in Ayacucho time and the mentor sees India time. University students often switch to one to one at the start of each semester, when their own timetable changes.