Calibrated Researcher
I am Ilhame Ait Lbachir, a computer science professor and AI researcher. I help researchers do work that holds up to scrutiny: stronger research, sharper papers, proposals, and talks, with a specialist’s eye for trustworthy, uncertainty-aware AI.
For PhD students and researchers who want work that holds up to scrutiny.
Synthetic scan. Illustrative, not a real patient image.
What I do
Not louder work. Stronger work. Here is where I go deep.
Methodology and experimental design that survive a hard read. How to frame a question, run experiments that mean something, and report results reviewers trust.
The craft of writing for acceptance. How a paper actually gets written, where proposals fall apart, and what reviewers look for before they say yes.
From my own research: uncertainty quantification, explainability, and calibration for medical imaging. How to reason about AI you can actually trust, made clear.
A named standard
The standard I hold myself to, and the one I teach. Confidence that matches the evidence, never more.
Every sentence earns its place. If the data does not back it, it does not ship.
Read your own work the way a hostile reviewer will. Find the holes before they do.
Say what you do not know. Calibrated confidence is more persuasive than false certainty.
Proof
I. Ait Lbachir, I. Daoudi, R. Nassih · International Conference on Intelligent Systems and Digital Applications (ISDA), IEEE
I. Samiry, I. Ait Lbachir, I. Daoudi, S. Tallal, S. Adil · International Work-Conference on Bioinformatics and Biomedical Engineering (IWBBIO), Springer
I. Ait Lbachir, I. Daoudi, S. Tallal · Multimedia Tools and Applications (vol. 80), Springer
Field notes
What a doctorate actually trains you to do, the three jobs of year one, and the habits that make the rest of the PhD easier.
7 min readPhD basicsTurn a vague topic into a question that is new, answerable, and defensible, with five tests and a four step method.
8 min readPhD basicsThe three pass method, a reviewer's checklist for judging evidence, and a note template that turns reading into writing.
8 min readTrustworthy AI made clearA plain map of uncertainty quantification: what it is, the main methods and their cost, how to test them, and what to report. Free PDF kit included.
9 min readTrustworthy AI made clearTwo kinds of "I am not sure": noise in the data, and gaps in the model. Plain examples, how to estimate each, and what to do with them.
8 min readGetting papers acceptedSort the comments, answer every one, say what changed and where, and disagree with evidence. A calm method for the rebuttal letter.
8 min readFree kit
The checks I run before any paper goes out, distilled from reviewing more than fifty of them. The things reviewers reject for, before they ever reach a reviewer.
A printable PDF. No email, no sign-up. Use it on your next paper and share it with your lab.
Download the checklist (PDF)Courses
Hosted on Udemy. Leave your email on the courses page and you will hear the day it opens.
The full workflow from question to accepted paper. Experimental design, honest evaluation, and writing that gets past reviewers.
About
I am a computer science professor and AI researcher working on AI you can actually trust. Calibrated Researcher is where I share how that work really happens. Not the polished version. The real one.
Read the full story