Most CVs flatten a career into a list. This one doesn’t. Every degree, role, thesis, and product I’ve shipped is placed once, by hand — then four lenses redraw the connections between them. Choose the thesis you want to read: a continuous engineering spine, a tightening ML loop, a fleet of shipped products, or a research compass spanning Bogotá and London.
Years
2018 → 2026
Arc
Bogotá → London
Stars
13
Lenses
04
Edges
20
How to read this
Pick a lens and the field redraws — the stars that belong to that reading stay lit, the rest dim. Hover a star to hear what it is. Click to pull the full chapter alongside. The URL follows along, so any particular reading is a link you can share.
The four readings
One career, four shapes.
The field of stars doesn’t move. Only the edges between them do. Each lens is an editorial claim about which story is worth telling — and every claim is a link you can share.
Lens I · Engineering
The Throughline
The continuous software-engineering spine.
The continuous software-engineering spine. A single line traced from first principles at Los Andes to production practice at Untap, with every role compounding the last.
Seven stars, six edges. Early ML forks off the undergrad — an F1 predictor at ~93% accuracy — and the chronology rejoins at Imperial before landing in a production-grade LLM surface in London.
BSc Computing
F1 Prediction
Bancolombia
Glamper
MSc Computing
Untap
Financial Extraction
7 stars · 6 edges
Lens II · ML Systems
The Loop
Research, extraction, retrieval, evaluation — and back again.
A closed loop of research, extraction, retrieval, and evaluation. Each node feeds the next; each measurement tightens the loop. The thesis is where the rigor originates, and every product here inherits it.
Five stars, five edges — the only lens that actually loops. Fine-tuned extraction at 90%+ accuracy feeds an agentic RAG + MCP layer; EvalLens holds both to account; LFM carries the same discipline into football.
HPC Regression Detection
Financial Extraction
Investor Intelligence
EvalLens
Large Football Model
5 stars · 5 edges
Lens III · Products Shipped
The Fleet
Named vessels, each with its own course.
A fleet of personal vessels — each its own artefact, each under its own flag. The portfolio is the harbour; the career graph and The Constellation are sibling visualisations of the same career, read differently.
Four stars, five edges. Every vessel traces back to the same harbour: rendonarango.com. Each is self-contained, deep-linkable, and shipped without a heavy framework substrate behind it.
rendonarango.com
Career Graph
EvalLens
Large Football Model
4 stars · 5 edges
Lens IV · Research & Learning
The Compass
The directions the work has pointed in.
A compass rose of formal study. Undergrad and postgrad anchors with a thesis on each end — race-winner prediction in Bogotá, HPC regression detection in London. Two different problems, the same evaluation discipline underneath.
Four stars, four edges. The shape is deliberate — two institutions, two theses, one methodology. Evaluation rigor is the line that runs through the middle.
BSc Computing
F1 Prediction
MSc Computing
HPC Regression Detection
4 stars · 4 edges
Measured
Rigor leaves fingerprints.
Evaluation discipline runs through the constellation from end to end — from a Formula 1 thesis in Bogotá to production LLM tooling in London. A few of the marks it left.
BSc thesis · F1 winner prediction
~93%
Winner-classification accuracy over 70+ seasons of Formula 1. Feature-engineered telemetry, weather, grid, and form signals, explained with SHAP.
Untap · financial extraction
90%+
Fine-tuned LLM pipeline pulling KPIs, targets, and cashflow movements out of dense financial documents into structured, reviewable data.
Untap · ingestion time
−80%
Client-side data ingestion time cut against the baseline manual process. Analysts shifted from data entry to decision-making.
MSc thesis · HPC triage
days → <2h
Variational-autoencoder anomaly detection over HPC performance counters, integrated into CI for Nektar++ — commit-level regression surfacing.
Editorial readings
04
Four lenses redraw the same field of stars — the engineering spine, the ML loop, the shipped fleet, the research compass.
Personal vessels
05
rendonarango.com, the Interactive Career Graph, EvalLens, the Large Football Model, and this constellation — each shipped under its own flag.
Colophon
If any of this looks like your problem.
I’m Simon — an ML systems engineer at Untap, building the edges you just traced. Interesting work, open problems, and the occasional football match are all good reasons to write.