-
Derivatives and gradients: the one calculus lesson ML actually uses
Machine learning can look like a huge collection of algorithms, architectures, and optimization tricks. Underneath most of them, however, sits one compact idea:
-
Weight initialization: why your deep net trains or dies before step one
A deep network does not begin learning from a neutral state. Before the optimizer takes its first step, the initial weights have already determined:
-
Model monitoring in production: drift, decay, and alerting
A conventional service usually fails loudly. It times out, returns an error, exhausts memory, or stops responding.
-
Vision-language models: teaching an LLM to see
A text-only language model receives a sequence of tokens and predicts which token should come next. A vision-language model extends that idea by converting images into representations that can partici
-
Hybrid search: combining BM25 keyword and dense vectors
Dense vector retrieval is excellent at matching meaning.
-
Fine-tuning vs LoRA vs prompting: choosing the cheapest path that works
The cheapest model-customization strategy is not the one with the smallest training invoice. It is the one that reaches the required quality level with the lowest total lifecycle cost.
-
Forecasting with XGBoost: turning a time series into a supervised problem without cheating
XGBoost does not know what a time series is.
-
RuleFit and skope-rules: mining crisp rules out of a trained ensemble
Tree ensembles are excellent at discovering nonlinear thresholds and feature interactions. A boosted classifier can learn that a large order is suspicious only when the account is young, or that a dis
-
Keeping an XGBoost model honest: drift monitoring, champion/challenger, and retraining cadence
An XGBoost model can remain perfectly healthy from an infrastructure perspective while becoming operationally wrong. The endpoint still returns 200 OK . Latency stays flat. CPU and memory look normal.
-
Reproducible XGBoost: seeds, pinned versions, and a validation protocol you can defend
A reproducible training run and a trustworthy validation score are related, but they are not the same guarantee.