Technology LearningArticles
Short pieces on the things readers get stuck on. Free to read.
List, tuple or dictionary? How to pick the right Python container
Python gives you three everyday ways to hold a group of values. Here is how to tell them apart and choose the right one without overthinking it.
for loop or while loop? A simple rule for Python
Both loops repeat work, so which one do you reach for? The answer comes down to one question you can ask before you write a single line.
Reading a file in Python without the three common mistakes
Opening a file is easy. The mistakes are always the same three, and once you know them you never trip on them again.
split() in Python: turning one string into many
The split method is how a single line of text becomes a list of separate values. It is small, it is everywhere, and it has two habits worth knowing.
SQL JOIN types explained: INNER, LEFT, RIGHT and FULL
A join stitches two tables into one result. The type you choose decides what happens to the rows that do not match, and that is the whole difference.
What is database normalization? 1NF, 2NF and 3NF in plain English
Normalization sounds like heavy theory. It is really one habit, applied three times: store each fact in exactly one place.
Primary key vs foreign key: what each one does
Two words that sound similar and do opposite jobs. One names a row, the other points at one. Get the pair right and your tables link up cleanly.
How a database index actually speeds up a query
An index is the reason a query on millions of rows can return in a blink. Here is what it really is, and why adding one everywhere is a mistake.
What is a DRG, and how is one assigned?
A DRG turns a whole hospital stay into a single payment category. Here is what it is, what feeds into it, and why coders sit at the centre of it.
How an ICD-10-CM code is built, character by character
An ICD-10-CM code is not a random string. Each position has a job. Learn the shape once and the codes stop looking like a foreign language.
Code first and use additional: the etiology and manifestation rule
Some conditions must be coded as a pair, in a set order. The code first and use additional notes are the manual telling you which comes first and why.
Principal diagnosis: the one rule that decides everything else
The definition is one sentence long. Almost every sequencing mistake comes from reading it too quickly.
What Z codes are for: coding when there is no disease
Not every visit is about an illness. Z codes exist for the reasons people meet the health system when the reason is not a disease at all.
The 7th character, explained with a broken wrist
Injury codes carry an extra character at the end that says which visit this is. Here is what it does and why the X is there.
Zero-shot vs few-shot prompting: what examples do to an answer
The fastest way to improve an AI answer is often to show it an example or two. Here is the difference between asking cold and asking with examples.
Reading an HL7 message for the first time
An HL7 v2 message looks like a wall of pipes and carets. It is actually a very simple shape, and you can learn it in ten minutes.
What an AI agent really is, and how it differs from a chatbot
The word agent is used loosely. The real difference is simple: an agent can take actions and decide its own next step, not just answer.
How to check an AI answer before you trust it
AI is confident whether it is right or wrong. A few quick habits let you catch the wrong answers before they cost you anything.
One email when a new book lands
No schedule, no filler. Just the new titles and the odd free chapter.