You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I build production systems that turn messy data into decisions. Two decades across logistics, health, and engineering taught me the value of lean models, clean code, and reproducible pipelines. Lately I’ve been shipping NLP and statistical modelling that helps teams reason about text and time series in real time.
10
+
I build production systems that turn messy data into decisions.
11
+
Across logistics, health and engineering I’ve focused on lean models, clean code, and reproducible pipelines.
12
+
Lately I’ve been shipping **sensor analytics, survival data tools, drift / anomaly detection**, and **LLM-powered reporting** that help teams reason about time series and text in real time.
14
13
15
-
<palign="center">
16
-
<img src="data_has_a_better_idea.png"
17
-
alt="Poster with the phrase 'Data has a better idea'"
18
-
title="Data has a better idea"
19
-
width="75%" />
20
-
</p>
14
+
-**smart-todo-action** — GitHub Action that extracts TODOs, groups by semantic labels / tags / metadata, and opens issues / changelogs.
68
+
-**navier-stokes-solvers** — CFD solvers for the 2D / 3D Navier–Stokes equations (finite-difference & spectral variants), with buildable CLI targets and basic tests.
69
+
-**heavytails** — Utilities for heavy-tailed modelling and inference (tail index estimation, Pareto-like fits, EVT diagnostics).
-**navier-stokes-solvers** — CFD solvers for the 2D/3D Navier–Stokes equations (finite-difference & spectral variants), with buildable CLI targets and basic tests.
Vector spaces and linear maps; matrices and determinants; eigenvalues / eigenvectors, diagonalisation; orthogonality, projections, Gram–Schmidt; least squares; SVD and PCA; numerical stability & conditioning; applications to optimisation and data science.
84
+
Syllabus: link · Slides (Beamer): link
85
85
86
-
### Teaching @ESMAD
87
-
-**Introduction to Logic & Set Theory (First Semester, 15 weeks)** — Logic (prop/FO), sets, induction, **differential & integral calculus**; notes + LaTeX.
88
-
-**Linear Algebra (Second Semester, 15 weeks)** — Vector spaces and linear maps; matrices and determinants; eigenvalues/eigenvectors, diagonalisation; orthogonality, projections, Gram–Schmidt; least squares; **SVD and PCA**; numerical stability & conditioning; applications to optimisation and data science.
89
-
Syllabus: _link_ · Slides (Beamer): _link_
90
-
-**NLP & LLM mini-workshops** — Prompt design, evals, lightweight retrieval, and report generation with structured → narrative transforms.
86
+
-**NLP & LLM mini-workshops**
87
+
Prompt design, evals, lightweight retrieval, and report generation with structured → narrative transforms.
91
88
92
89
### Seminars & Workshops
93
-
-**Data Science Seminars** — End-to-end ML pipelines, feature engineering for time series, evaluation under drift, MLOps (CI/CD, data/versioning), and reproducible research practices.
94
-
Slides: _link_ · Notebooks: _link_
95
-
-**Sensors & Dashboards** — IoT data ingestion (MQTT/Kafka), time-series storage (InfluxDB/Parquet), streaming analytics (Flink), and dashboards (Grafana/Plotly/Dash) with alerting & anomaly detection.
96
-
Slides: _link_ · Demo repo: _link_
97
-
-**Applications of Matrices to Computational Graphics** — Linear transforms in 2D/3D, homogeneous coordinates, rotations (Euler vs. quaternions), camera models & projections, shading basics; **SVD/PCA** for geometry processing.
98
-
Slides: _link_ · Code samples: _link_
90
+
91
+
-**Data Science Seminars**
92
+
End-to-end ML pipelines, feature engineering for time series, evaluation under drift, MLOps (CI/CD, data / versioning), and reproducible research practices.
93
+
Slides: link · Notebooks: link
94
+
95
+
-**Sensors & Dashboards**
96
+
IoT data ingestion (MQTT / Kafka), time-series storage (InfluxDB / Parquet), streaming analytics (Flink), and dashboards (Grafana / Plotly / Dash) with alerting & anomaly detection.
97
+
Slides: link · Demo repo: link
98
+
99
+
-**Applications of Matrices to Computational Graphics**
100
+
Linear transforms in 2D / 3D, homogeneous coordinates, rotations (Euler vs. quaternions), camera models & projections, shading basics; SVD / PCA for geometry processing.
101
+
Slides: link · Code samples: link
99
102
100
103
### Selected Writings / Demos
101
-
-**Streaming analytics with Iceberg + Flink + DynamoDB** — Architecture notes and example pipelines.
102
-
-**Robust regression with IRLS** — ψ-functions, influence diagnostics, and uncertainty reporting.
103
-
-**Time-series anomaly detection** — EWMA variants, adaptive σ, and change-point alerts for sensors.
104
+
105
+
- Streaming analytics with Iceberg + Flink + DynamoDB — architecture notes and example pipelines.
106
+
- Robust regression with IRLS — ψ-functions, influence diagnostics, and uncertainty reporting.
107
+
- Time-series anomaly detection — EWMA variants, adaptive σ, and change-point alerts for sensors.
Thanks for visiting! I’m keen to partner with data enthusiasts, researchers, and product teams. Browse my projects or get in touch—happy to explore ideas and build useful things together.
0 commit comments