Portfolio · 2026 · v1.0

The space between
what is and what could be...
that's where I build.

Finance background. Builder's instinct. Still learning. Two years in the workforce, building tools that save time and surface the right information — for myself, for colleagues, and for whoever picks it up after me.

BackgroundFinance × Data
ApproachAdopt · Build · Improve
TargetAnalyst · Fintech · Data
Index— · —

The work

02 / Proof

Six projects built from real problems. Each one designed so someone other than me can use it, understand it, and build on it. Domain knowledge applied. Tools chosen to fit the problem.

02 / Finference · PlatformComing soon

Finference, finance inference.

A coming equity research web app for scenario models, real market data, fundamentals, technical indicators, local AI inference, and Discord signal alerts.

Bull case+18.4%
Base case+6.7%
Bear case−9.2%
Fundamentals82
Technical64
Noise cut71%
PythonSQLLocal AIAPIsDiscord Bots
03 / Case Study · InfraComing soon

Alberta hyperscaler case study.

A reporting and modeling project on hyperscaler demand, Alberta power economics, infrastructure constraints, and the investment case behind site selection.

·
Load
Power
Capex
IRR
Risk
1
120MW
$54
$1.2B
14.8%
Grid
2
240MW
$49
$2.1B
17.2%
Permit
3
360MW
$46
$3.0B
18.6%
Water
4
480MW
$52
$4.4B
−210bps
Queue
Power BIExcelModelingReporting
04 / Sports · ModelsComing soon

Sports analyst, game inference.

A multi-sport analysis engine for game previews, real-time betting odds, scraped stats, and simple math models that explain where the market may be off.

1# odds + team form + injury context
2def price_game(market, model):
3 odds = scrape_live_odds(market)
4 edge = model.probability() - odds.implied
5 return explain(edge, confidence=0.72)
6
7# output: analysis, not gambling advice
PythonSQLWeb ScrapingOdds APIs
06 / Power BIComing soon

Operating cohort lens.

Retention, ACV, and margin drift surfaced in one view — designed so a commercial team can spot the trend without opening a spreadsheet.

Cohort '2387%
Cohort '2482%
Cohort '2594%
ACV$11.4k
GM71.2%
Net Rev Ret118%
Power BICohortsSQL

Tools & approach

04 / Stack

Tools are chosen to fit problems, not the other way around. Finance-native first. Code where it adds leverage. AI where it closes the gap — then pressure-tested against what the domain actually requires.

Palette

BG0B0B0D
SOFT111114
SOFT-216171C
LINE22232A
SIGNALFF6A1F
INKF4F1EA
DIMA8A59C
DIMMER6B6A64
GLOWFF8A4D
DUAL

One signal colour. Two surfaces of black. One paper white. Everything else is restraint.

Type

Fraunces — emphasis.
Display · variable serif · 80% of personality
Inter — neutral, 300/400/500/600.
Body · interface · captions
JetBrains Mono — data.
Mono · tickers · eyebrow labels

Tools

01Excel · Power QueryDomain fluent
02Power BI · DAXDomain fluent
03Python · pandasBuilt & shipped
04SQL · DuckDBBuilt & shipped
05Next.js · TSThis portfolio
06LaTeX · PandocDeterministic docs