IB Math Applications and Interpretation SL · Online · One-to-one
IB math AI SL tutor for students who need interpretation, not repetition
A private IB math AI SL tutor for Diploma students taking Applications and Interpretation at Standard Level. One-to-one, online, worldwide. Built for students who need modelling judgement, statistics fluency and confident calculator use — not another explanation of the textbook. Application-only.
AI SL does not ask whether your child can execute a procedure. It asks whether they can read an unfamiliar context, choose a sensible model, run it on the GDC, and say what the result means in the language the mark scheme rewards. Students who memorised their way through IGCSE find this brutal — because the skill being tested is precisely the one they never had to build.
My work as an IB maths Applications and Interpretation SL tutor starts from a different place: a diagnosis of the exact step where your reasoning broke. Not "statistics is weak" — the specific decision, in a specific question type, where you stopped thinking and started guessing. From there we rebuild on a fixed weekly rhythm, every session recorded and archived into a searchable library you keep. This page covers AI SL; the full IB Math tutoring practice covers all four courses. Based in Milan, working online with students worldwide. The exact investment is set at the intake conversation; places are application-only.
Unfamiliar contexts: you chose AI for "easier" maths, and got contexts you don't recognise
AI SL is not easier; it is different. A question about fish stocks, phone battery decay or hospital waiting times hands you a scenario before it hands you any mathematics, and the first marks go to students who can translate that scenario into a model. If you were trained to wait for the question to tell you which formula to use, AI SL feels unpredictable. The fix is not more content revision — it is practising the translation step itself, systematically, across the contexts the exam actually uses.
GDC technique: quietly costing you marks in both papers
Both AI SL papers allow the GDC, and the course assumes fluency: regression setups, normal and binomial distributions, graph analysis, solving from intersections and tables. A student who types the right numbers into the wrong regression, or who cannot extract what the calculator output means, loses marks that never look like "calculator errors" — they look like wrong answers. GDC fluency is trainable, and in AI SL it is worth more marks per hour of practice than almost anything else.
Statistics interpretation: the maths is easy, the marks still don't come
AI SL marking schemes reward precise interpretive language. Saying a correlation is "good" earns nothing; stating that r = 0.87 indicates a strong positive linear relationship, and that this does not by itself imply causation, earns the marks. Many capable students bleed points across hypothesis testing, sampling and probability questions simply because nobody taught them the vocabulary the examiner is listening for. This is one of the fastest gains available in the course.
Exploration structure: a topic with no real question, or a question with no structure
AI SL Explorations live or die on a clear real-world question and honest reflection. The most common failure is a beautiful piece of data analysis that never answers anything, or a personal topic with mathematics too shallow for the criteria. Twenty percent of the final grade sits here. I help students sharpen the question, choose mathematics appropriate to SL, and write reflection that reads as genuine engagement rather than filler — within the IB's academic honesty rules, which I take seriously.
Recovery from a low level: not polish, an actual rebuild
Some students arrive at AI SL already behind: a Grade 2 or 3, gaps from earlier years, and rising panic. This page exists partly for you, because the strongest documented case on this site is exactly this situation — a student who began her Diploma at Grade 3 in AI SL and finished with Grade 7 in both final exam papers. Recovery at this scale is not typical and is never promised. But it is possible, and it starts with an honest diagnosis rather than a crash course.
The AI SL syllabus, covered properly
What an AI SL tutor online should actually cover
Applications and Interpretation SL is organised around number and algebra, functions, geometry and trigonometry, statistics and probability, and calculus — but the exam's centre of gravity is modelling, data and interpretation. Two papers, both with GDC allowed, no Paper 3. Coverage below follows the current syllabus (first assessment 2021), with sessions weighted towards where your diagnosis shows the marks are being lost. The full specification is on the IB's mathematics courses page.
- Number and algebra: sequences and series, financial mathematics (compound interest, annuities, amortisation) — the most heavily weighted topic in AI SL Paper 2 — laws of exponents and logarithms, approximation and error
- Functions: linear, quadratic, exponential and polynomial models; modelling real situations; inverse functions; using the GDC to analyse graphs and intersections; piecewise functions in step-context problems (tax brackets, postal rates)
- Geometry and trigonometry: right and non-right triangle trigonometry, sine and cosine rules, sectors, Voronoi diagrams (construction, nearest-neighbour queries, the toxic waste dump problem), volume and surface area in context
- Statistics: sampling methods and when each applies, presentation of data, measures of central tendency and spread, correlation and linear regression (including interpreting r and least-squares lines in the exact language mark schemes expect), reliability and validity of data
- Probability: combined and conditional probability, tree and Venn diagrams, discrete random variables, binomial and normal distributions — with correct calculator setup and interpretation
- Calculus: introduction to differentiation, gradients of curves, increasing and decreasing functions, basic optimisation; integration as area and accumulation
- The Mathematical Exploration: question design, appropriate SL-level mathematics, criterion-by-criterion drafting feedback, honest reflection
- GDC fluency drills: weekly timed reps — ten regression setups in ten minutes, five normal-distribution answers using `normalcdf`/`invNorm`, five binomial-distribution answers using `binompdf`/`binomcdf` — until the calculator stops being a black box
- Exam technique: both AI SL papers under timed conditions, mark-scheme interpretive-language training, structured review of every wrong answer
The method
Diagnosis first. Then a fixed rhythm that compounds.
We begin by finding the exact step where your reasoning broke — not the topic, the step. Sessions then run on a fixed weekly rhythm, each one opening with last week's wrong answers, because errors you have already made are the cheapest source of marks. We work on Zoom with a shared Excalidraw whiteboard and your GDC emulator where useful. Every session is recorded, summarised, and archived into a searchable library you keep, so revision before Paper One and Paper Two is retrieval, not archaeology.
Documented, not promised
From Grade 3 to two Grade 7 papers: the Aline Pascarelli case
Aline Pascarelli arrived at the start of her Diploma Programme with a Grade 3 in IB Maths AI SL and finished with a Grade 7 — a 7 in each written paper. Her words: "As a student who moved from a Level 3 in IB Maths AI SL at the beginning of my IB course to obtaining a Level 7 in my final exam papers, I would highly recommend Renzo… I am confident I would not have been able to obtain this grade without his help." Public Google review.
Recovery from a low level is possible, and it is easier the earlier it starts — a diagnosis in the first term of DP1 leaves the most runway. I share documented cases as precedent, not promise: your result depends on your starting point, your work between sessions, and the time available.
AI SL rewards judgement. Judgement can be trained.
The IB math AI SL papers do not ask who has memorised the most procedures. They ask who can read an unfamiliar context, choose a model, run it cleanly on the calculator, and say what the answer means. That is a trainable skill — and the training works best one-to-one, on a fixed rhythm, with every wrong answer examined rather than skipped. If that is the kind of help you are looking for, the next step is an application.