Saarland already puts more Informatik on the timetable than most of Germany, so a Saarbrücken teenager meets programming at school from grade 7. What school cannot do is follow one learner at their own speed, or show a parent how a model learns. This page covers who our live classes in English are for, from a six year old to an adult, what Saarland schools teach and when, the AI and puzzle contests coming up, and a project on the Ensheim weather station where a model gets better the moment we give it a fair ruler.
Teaching live online since 2020 · 10,000+ learners worldwide · every class in English, one regular mentor
In short
Modern Age Coders teaches coding and AI to learners in Saarbrücken live online in English: Scratch and Python and AI for Kids from age 6, Python and machine-learning work for teenagers, and AI and ML courses for adults, either in a class of fifteen to twenty or alone with the mentor. One regular mentor matches each learner's pace and accent. Saarland schools teach Informatik from grade 7; our classes add depth, projects and a learner's own speed.
Three starting points
Which course fits which learner in Saarbrücken?
One course per stage of life: a child who wants to make things, a teenager who wants to know how AI really works, and an adult who wants machine learning they can use at work.
For anyone in the city who wants to build with code rather than only use apps: primary children, pupils at Gemeinschaftsschule or Gymnasium, students, and adults changing direction.
Children, 6 to 12
Younger children start with Scratch blocks and stories, so nothing has to be typed in the first weeks. From about nine, they write real Python and meet AI as a set of ideas: sorting, guessing, learning from examples. No adult chatbot accounts are needed for any of it.
Teenagers, 13 to 18
A Saarland pupil already has Informatik lessons from grade 7. Our teenage courses go further and faster: Python projects, data, and machine learning written step by step, which is also the kind of work the Bundeswettbewerb Künstliche Intelligenz asks for.
Adults
Parents, university students and professionals take one-to-one or group courses in Python, data science and AI. Many want to understand the tools their employer is bringing in; others want a project portfolio. The mentor plans around evenings and weekends.
A first course for each learner (our recommendation)
Learner
Typical start
Moves on to
Child aged 6 to 8
Coding for Kids (Scratch first)
Python and AI for Kids
Child aged 9 to 12
Python and AI for Kids
Maths through coding, or AI literacy
Pupil in grades 7 to 10
Python for Teens
Informatik Companion, Klasse 5 to 10
Pupil in the Oberstufe
AI and Machine Learning for Teens
Informatik Abitur in English
Adult beginner
Python Masterclass
Data Science or the AI and ML Masterclass
The same family can have three learners on three different courses, each with their own mentor and their own times. A child who started at seven and a parent who started at forty are both judged on the same thing: what they can build and explain at the end of a month.
Saarland schools
How much Informatik does a Saarbrücken pupil get at school?
More than in most Länder, and it starts in grade 7. The figures below come from the Gesellschaft für Informatik and the Kultusministerkonferenz, read on 2 November 2026.
Compulsory from grade 7
According to the Gesellschaft für Informatik's Informatik-Monitor 2026/27, Informatik is compulsory at Gemeinschaftsschulen and at the nine-year Gymnasium in Saarland: two lessons a week in grades 7 and 8, and one lesson a week in grades 9 and 10.
On the national page of the same monitor, Saarland is one of only three Länder, with Mecklenburg-Vorpommern and Thüringen, that already meet the medium-term target of six compulsory hours recommended by the KMK's Standing Scientific Commission for 2030.
What that leaves open
Six hours across four school years is a solid start, but it is shared between a whole class at one speed. Before grade 7 there is no fixed computing lesson at all, which is the gap younger Saarbrücken children fall into.
The Gymnasium is also changing: the monitor notes that Saarland moves to the nine-year G9 route by 2026/27. In the Oberstufe, Informatik is a course a pupil chooses; nationally, about 15 percent of pupils in the Qualifikationsphase take it.
Informatik in a Saarland school career (Gemeinschaftsschule and G9 Gymnasium)
Grades
Computing on the timetable
1 to 4
Grundschule: no fixed Informatik lesson
5 and 6
No fixed Informatik lesson
7 and 8
Informatik, two lessons a week, compulsory
9 and 10
Informatik, one lesson a week, compulsory
Oberstufe
Informatik as a chosen course, where the school offers it
Sources: Gesellschaft für Informatik, Informatik-Monitor 2026/27, Saarland page and Länder overview; read 2 November 2026. In 2026 the KMK adopted an update of its framework Bildung in der digitalen Welt that adds AI-related competences. We are not connected to the Gesellschaft für Informatik, the KMK or the Saarland ministry, and we do not award school grades: the school course is the one that counts, and ours runs alongside it.
Saarland school holidays 2026/27 (KMK)
Holiday
Dates
Herbstferien
5 to 16 October 2026
Weihnachtsferien
21 to 31 December 2026
Winterferien
8 to 12 February 2027
Osterferien
30 March to 9 April 2027
Sommerferien
28 June to 6 August 2027
Source: Kultusministerkonferenz, Ferienkalender 2026/27, Stand 9 October 2025. Classes carry on through school holidays; families who travel simply pause and pick up where they stopped.
Taught in English
Are the classes really in English, and does that work for a German speaker?
Yes, every lesson is live and in English, and for most learners in Saarbrücken that turns out to be an advantage rather than a hurdle.
No German lessons, said early
We teach in English only. A family that needs teaching in German should know that before the first class, so we say it here and again in the questions below. Mentors are based in India and teach over live video.
The mentor adapts
Mentors slow down, rephrase, and wait. A nine year old who answers half in German is understood and handed the English word; an adult who wants technical vocabulary gets it written in the chat. Accent and pace are matched to the learner, not the reverse.
Programming is already English
Python is written with words like if, while, print and return. The error messages, the documentation and most of the help online are in English too, so a learner who practises the language of programming in class reads it more easily everywhere else.
Saarbrücken sits a few kilometres from the French border, and many families here already live with more than one language. A lesson in English adds one more, in a subject where the vocabulary is small and the screen does much of the explaining.
A real Saarbrücken project
Can a model guess tomorrow's temperature at Ensheim from days that looked like today?
This is the project our teenagers and adults build in their first machine-learning month. It uses one weather station inside the city, a simple idea called nearest neighbours, and one mistake that almost everyone makes on the first try.
The data
The Deutscher Wetterdienst runs station 4336, Saarbrücken-Ensheim, at the airport in the south-east of the city, 319 metres above sea level. Its Climate Data Center publishes the daily observations from 1951 to 2025 under the Creative Commons BY 4.0 licence. We downloaded the file on 3 November 2026.
We keep four columns from 31 December 1990 onwards: the date, the day's highest temperature, the mean air pressure and the relative humidity. That leaves 12,783 days with all three values present, saved as ensheim_daily.csv.
The idea
To guess tomorrow, find the 20 past days that looked most like today and average what happened the day after each of them. Nothing is hand-written about weather; the model only remembers and compares. Each day is described by four numbers: the day of the year, its highest temperature, how much the pressure changed since yesterday, and the humidity.
The model learns from 1991 to 2024 and is tested on 2025, a year it has never seen. It has to beat the laziest guess there is: tomorrow will be as warm as today.
Version 1: nearest neighbours in raw unitsPython
import csv
from datetime import date, timedelta
days = {}
with open("ensheim_daily.csv", encoding="utf-8") as source:
for record in csv.DictReader(source):
day = date.fromisoformat(record["date"])
days[day] = (float(record["tmax"]), float(record["pressure"]), float(record["humidity"]))
examples = []
for day in days:
before, after = day - timedelta(days=1), day + timedelta(days=1)
if before in days and after in days:
tmax, pressure, humidity = days[day]
features = [day.timetuple().tm_yday, tmax, pressure - days[before][1], humidity]
examples.append((day.year, features, days[after][0]))
train = [e for e in examples if e[0] <= 2024]
test = [e for e in examples if e[0] == 2025]
def difference(i, a, b):
if i == 0: # day of the year goes round in a circle
return min(abs(a - b), 365 - abs(a - b))
return a - b
def predict(features, units, k=20):
scored = []
for _, other, tomorrow in train:
dist = sum((difference(i, features[i], other[i]) / units[i]) ** 2 for i in range(4))
scored.append((dist, tomorrow))
scored.sort()
return sum(t for _, t in scored[:k]) / k
def error(pairs):
return sum(abs(guess - real) for guess, real in pairs) / len(pairs)
print("Training days:", len(train), " Test days:", len(test))
print("Same as today:", round(error([(f[1], real) for _, f, real in test]), 2), "C")
raw = [(predict(f, [1, 1, 1, 1]), real) for _, f, real in test]
print("20 neighbours, raw units:", round(error(raw), 2), "C")
It prints Training days: 12413 Test days: 364, Same as today: 2.59 C and 20 neighbours, raw units: 2.53 C.
The model is only six hundredths of a degree better than copying today. The reason is in how distance is measured. In raw units, a gap of 30 days in the calendar counts the same as a gap of 30 degrees, and a gap of 13 humidity points the same as 13 degrees. So the model mostly looks for days at the same time of year with similar humidity, and hardly notices whether today was hot or cold.
Version 2: give every feature the same rulerPython
from statistics import pstdev
units = [pstdev([difference(i, e[1][i], 1 if i == 0 else 0) for e in train]) for i in range(4)]
print("Spread of each feature:", [round(u, 1) for u in units])
scaled = [(predict(f, units), real) for _, f, real in test]
print("20 neighbours, scaled:", round(error(scaled), 2), "C")
Run after Version 1. It prints Spread of each feature: [52.7, 8.5, 5.0, 13.5] and 20 neighbours, scaled: 2.39 C.
Dividing each difference by that feature's typical spread turns every feature into the same unit: "how unusual is this gap". Now a day that was 8.5 degrees warmer is as far away as one that sits 52.7 days apart in the calendar. The average error falls to 2.39 degrees, a fifth of a degree better than copying today. Nothing about the data changed. Only the ruler did.
Average error on the 364 test days of 2025 (our calculation)
Method
Average error, degrees C
Tomorrow will be like today
2.59
20 neighbours, raw units
2.53
20 neighbours, every feature scaled
2.39
Data: Deutscher Wetterdienst, Climate Data Center, daily climate data (KL), station 4336 Saarbrücken-Ensheim, licence CC BY 4.0, downloaded 3 November 2026. The column selection, the training and test split and all errors are our own. Two days before 2025 lack one of the values and are left out; 2025 contributes 364 test days because the last day of the year has no tomorrow in the file.
Children, 9 to 12. Write five cards with a day's temperature and the next day's. For a new card, find the most similar old card and copy its answer. That is a nearest neighbour.
Teenagers. Run Version 1, then change the humidity column to fractions (0.75 instead of 75) and run it again. Explain why the answer moves.
Adults. Try k of 5, 20 and 80 after scaling, and plot the error. Then leave out one feature at a time and find the one the model needs most.
Everyone. Write in one sentence why "it will be like today" is so hard to beat for the weather, and what kind of day it gets most wrong.
Contests this school year
Which coding and AI contests can a Saarbrücken pupil enter in 2026 and 2027?
Three national ones, from a short puzzle quiz to a full machine-learning project. We are not connected to any of their organisers.
Dates as published by the organisers
Contest
Who
Key date
Informatik-Biber (BWINF)
Grades 3 to 13, no prior knowledge
Biberwochen 9 to 20 November 2026, through the school
Bundeswettbewerb Künstliche Intelligenz
Sekundarstufe I and II, alone or in teams
Final in Tübingen, 13 November 2026
Känguru der Mathematik
Grades 3 to 13
8 April 2027, through the school
The AI contest
The Bundeswettbewerb Künstliche Intelligenz is run by the Tübingen AI Center. A team hands in written documentation, runnable code and a short video pitch, and the project must use machine-learning methods. Projects may be submitted in German or in English.
The 2026 round closed for projects on 20 September 2026 and ends with the final in Tübingen on 13 November 2026. A pupil who wants to enter the next round can start building now; a project like the Ensheim model is the right size for a first attempt.
The puzzle contests
The Informatik-Biber is a free online quiz of logic puzzles that schools run in a lesson. In 2026 the Biberwochen are 9 to 20 November. Ask your child's teacher whether the school takes part.
In class we never use real contest tasks. We write our own problems in the same spirit, label them as ours, and point families to the organisers' own free archives.
Class times
When do classes run in Saarbrücken time?
Late afternoon and early evening suit most learners here, and that is evening in India, where the mentors are. Times are agreed with you; there is no fixed timetable.
Four sample Saarbrücken slots, read on a clock in India
Saarbrücken
India, winter (to 28 March 2027)
India, summer (from 28 March 2027)
13:45
18:15
17:15
16:45
21:15
20:15
Saturday 08:30
13:00
12:00
Sunday 16:00
20:30
19:30
How the gap works: German law (the Einheitenzeitgesetz) sets MEZ, one hour ahead of UTC, for the winter months and MESZ, two hours ahead, for the summer. The clocks went back on 25 October 2026 and go forward on 28 March 2027. India never changes its clocks and sits five and a half hours ahead of UTC, which puts a mentor four and a half hours ahead of Saarbrücken until the spring change and three and a half hours ahead after it. The slots above illustrate the conversion; none of them is a fixed class.
Learners from Völklingen, St. Ingbert, Sulzbach or Dudweiler join the same way, because everything happens online. When you book, the time is shown in your own clock, and the mentor sees it in theirs.
Progress
What can a learner do after each stage?
Our own stages, the same for every age; only the pace differs.
Stages of coding and AI (ours)
Stage
The learner can
1. Instructions
Give a computer exact steps and fix them when they go wrong
2. Programs
Write Python with loops, conditions and functions
3. Data
Read a real file, clean it and describe it in numbers
4. Models
Build a simple model by hand and test it on unseen data
5. Judgement
Say when a model is wrong, and why, before anyone else does
For parents
The Ensheim project sits at stage 4. A learner who has done it knows that a model is only as fair as the way it measures similarity.
Reports to you describe what was built and explained, not hours online.
For learners
Before you trust a distance, ask what unit each part of it is in.
A model that cannot beat a lazy guess has not learned anything yet.
More courses
Nine more courses for learners in Saarbrücken
All taught live in English, with one regular mentor, in a group or one to one.
A weather forecasting site with live conditions for any location.
by Krish
Fees
Fees for learners in Saarbrücken
A month buys eight live classes. The figure below is a euro estimate based on the European Central Bank reference rate; the actual charge goes through in US dollars, which your own bank changes into euro, so expect a few cents either way on the statement.
Google reviews from our students and their parents, unedited.
★★★★★
"The one step solution for my son. Modern Age Coders make learning coding so simple that kids love it. The teachers explain complex concepts clearly with practical exercises and interactive content."
Ria Mukherjee
Parent
★★★★★
"Modern Age Coders has been a game-changer for me. I struggled to grasp IT concepts and coding before joining, but their classes transformed everything. I can now confidently write complex programs with ease."
Samriddha Mondal
Student
★★★★★
"One of the most wonderful education centres out there. Education is not limited to school syllabus but focuses on skill development."
Vansh Agarwal
Student
★★★★★
"My child Dhairya is really enjoying the Modern Age Coders classes. This is his first online class and he eagerly looks forward to it. I can already see his improvement, and the teachers are very cooperative."
Sonam Oswal
Parent of Dhairya
★★★★★
"Modern Age Coders have wonderful teachers who teach in a clear, easy and practical way. The teacher boosts students' confidence and inspires them to learn without hesitation."
Sonu Goyal
Parent
★★★★★
"I highly recommend this computer coding class! The teachers are incredibly knowledgeable and passionate about coding."
Ritu Kedia
Parent
Questions from Saarbrücken
Coding and AI classes in Saarbrücken: questions and answers
Are the classes in German?
No, never. Lessons happen live in English, and only in English. Our mentors work from India; they slow down, choose plain words and get used to each learner's accent, which is why a ten year old with school English and an adult who learnt it long ago can both keep up.
Who can join coding and AI classes in Saarbrücken?
Children from 6, teenagers and adults. Children start with Scratch or Python and AI for Kids, teenagers with Python or AI and Machine Learning for Teens, and adults with the Python Masterclass or the AI and ML Masterclass.
Do Saarland schools already teach Informatik?
Yes. The Gesellschaft für Informatik reports that Informatik is compulsory at Gemeinschaftsschulen and the nine-year Gymnasium in Saarland, two lessons a week in grades 7 and 8 and one lesson a week in grades 9 and 10. Our classes run alongside school; the school course is the one that is graded.
When are classes in Saarbrücken time?
Weekdays or weekends, flexible: choose your batch, or for a very personalised plan choose 1-on-1 at your convenience, weekend mornings or weekdays. Most learners here pick a late afternoon, an early evening or a weekend morning.
Does my teenager need an AI chatbot account for these classes?
No. Nobody under 18 is asked to open one. Claude may only be used by adults, and ChatGPT requires a parent to agree for anyone under 18, so our teenage courses avoid both. Pupils write small models of their own instead, which shows them what a chatbot is doing far better than prompting one.
Can you help with the Bundeswettbewerb Künstliche Intelligenz?
We teach the skills a project needs: Python, data, machine learning and honest testing. The BWKI accepts projects in German or English. The 2026 final is in Tübingen on 13 November 2026, and the work for each entry must be the team's own.
Is the Ensheim weather project real data?
Yes. It uses the daily observations of Deutscher Wetterdienst station 4336, Saarbrücken-Ensheim, published under the CC BY 4.0 licence and downloaded on 3 November 2026. The model, the test on 2025 and the error figures are our own.
How many learners share a class?
A group class has between fifteen and twenty people, put together by age for children and by level for teenagers and adults. Someone who prefers undivided attention can take the one-to-one plan, with two private lessons each week.
How much does it cost, and in which currency?
There is one monthly fee. The page displays it in euro, worked out from the reference rate of the European Central Bank; the card is then charged in US dollars, and your bank turns that into euro on the statement, so a small difference is normal. Your first class costs nothing.
Can an adult with no coding background start with AI?
Yes, but through Python first. The AI and ML Masterclass begins with the basics of programming, then data, then models. Most adults are writing and testing their own small model within the first two months.
Read next
Related pages
Further reading for learners in Saarbrücken and Saarland.
In the first class the mentor gives a short problem, watches how the learner approaches it, and checks how comfortable they are in English. Afterwards you get a suggestion for the course and the group that fit.
Curious before booking? how we teach describes a live class, and all courses has every syllabus.
Live online classes with mentors in India, in English only. Write to us on WhatsApp in English for the quickest answer.
Want a class sooner? Book a priority demo
A priority demo is a full class of about 45 to 60 minutes. After the class the mentor sends a written skill report and a personal learning roadmap on WhatsApp. If you enrol, the demo fee is adjusted against your first month's fee. Share your demo Order ID with us on WhatsApp and we take it off.
