Finance, data, and commodity trade between Boston and Addis Ababa
About.
Mickias Ambaye
Boston, MA & Addis Ababa
Where I'm from
Addis Ababa, originally. Moved to Boston in 2021 for school and have been splitting time between the two cities ever since. The dual-country thing isn't a backdrop, it's most of why I think the way I do.
What I'm doing
Running finance for my family's coffee export business while working through the CQF in the evenings. The first explains why I started looking at risk seriously. The second is the part I find genuinely beautiful.
What I care about
Building things that work. A pricing model, a coffee shipment, a website. The satisfying ones are the ones where the math and the trade are the same problem from different angles.
Mickias Ambaye
Boston, MA & Addis Ababa
Where I'm from
Addis Ababa, originally. Moved to Boston in 2021 for school and have been splitting time between the two cities ever since. The dual-country thing isn't a backdrop, it's most of why I think the way I do.
What I'm doing
Running finance for my family's coffee export business while working through the CQF in the evenings. The first explains why I started looking at risk seriously. The second is the part I find genuinely beautiful.
What I care about
Building things that work. A pricing model, a coffee shipment, a website. The satisfying ones are the ones where the math and the trade are the same problem from different angles.
Quantitative Research.
Derivatives pricing, risk modeling, and stochastic calculus
Pricing
Exotic Options Lab
Asian, barrier, lookback, and American-exercise options under GBM. Monte Carlo with control variates and Longstaff-Schwartz regression.
Certificate in Quantitative Finance (CQF)
In ProgressAdvanced program covering derivatives pricing, risk management, stochastic calculus, numerical methods, and machine learning in finance.
Advanced Option Pricing Models
An interactive pricing studio covering Black-Scholes-Merton, Heston stochastic volatility (via the Carr-Madan Fourier transform), and Monte Carlo simulation with antithetic variates. Live Greeks surfaces, 3D implied-volatility skew, and a companion review paper synthesising CQF Module 3.
Financial Transmission Rights Pricing
Most option pricing models assume the underlying is a stock or a commodity. This paper takes the Black-Scholes framework and adapts it for Financial Transmission Rights, where the underlying is congestion risk on a power grid. Includes jump-diffusion models, seasonality adjustments, and interactive LMP charts.
Advanced ML Trading System
Coming SoonImplementation of methods from Lopez de Prado's Advances in Financial Machine Learning. CUSUM filters for event detection, triple barrier labeling, meta-labeling for signal confidence, purged K-fold cross-validation, and Kelly criterion for position sizing. Built in Python with XGBoost.
Analytics Portfolio.
Financial modeling, machine learning, and data-driven research
Commodity Risk Lab
Live VaR backtesting (Kupiec POF + Christoffersen independence), stylized-facts diagnostics, and side-by-side Gaussian / Cornish-Fisher / Filtered Historical Simulation method comparison on six commodity tickers. Data pipeline: Supabase + daily GitHub Actions ETL.
Spotify Popularity Time-Series Forecasting
Apply the same ARIMA / GARCH stack quants use for asset returns to a non-financial domain: track popularity on Spotify. Built on 586,672 tracks across 72 years (1950-2021). Includes a procedural Web Audio synthesiser that lets you hear what 'high energy, low valence' actually sounds like.
Coffee Futures Snowflake Warehouse
A production-style Snowflake data warehouse for coffee futures research. Galaxy schema (fact constellation) with 8 fact tables (~1.2M rows total) and 19 dimension tables stitched together from FMP, FRED, Open-Meteo, and the CFTC COT report. Designed to make a coffee thesis queryable in one join.
Eli Lilly DCF Valuation
DCF valuation of Eli Lilly (NYSE: LLY) using FCFF and FCFE with pipeline-adjusted revenue projections. Target price $627 against reference $579 with full scenario analysis and 9-peer relative valuation.
MNIST: MLP vs CNN
Technical comparison of a Multilayer Perceptron and Convolutional Neural Network on MNIST digit classification. Includes 3Blue1Brown-style architecture visualizations, convolution operation animations, and analysis of why spatial awareness matters.
Customer Segmentation Clustering
K-Means and hierarchical clustering on mall customer data. Identifies 5 spending behavior segments with ARI = 0.942 agreement between methods. The unsupervised methodology generalises directly to regime detection and factor-cluster portfolio construction on the quant side.
Also worth a look
3 side projectsUrban Heat Island ML Explorer
Interactive satellite ML classification explorer covering Rio de Janeiro, Santiago, and Freetown. Classifies urban heat island zones using the features extracted by UHI-Pipe.
Experience.
Work I've done in finance, trade, and operations
Chief Financial Officer
2024 to Present · Washington, DC
Run the financial side of an Ethiopian-American specialty-coffee export business — pricing, risk, working capital, and the quant models that drive the buy/hold/sell calls.
- ▸Scaled exports to roughly $5M in annual revenue (about 30 containers in FY2025, over 50 cumulatively in two years) by driving pricing and inventory decisions grounded in data-driven margin and demand analysis
- ▸Build quantitative models for the business: an inventory-hold profitability model for buy/hold/sell timing, plus volatility forecasting of coffee prices, operational-demand forecasting, and tariff-impact modeling to guide pricing, procurement, and inventory decisions
- ▸Manage net working capital and cash flow across AP/AR for both Muda and its Ethiopian vendor (HAI), coordinating the financing and operational needs of an intercompany supply chain spanning the US and Ethiopia
- ▸Own banking relationships and trade finance (letters of credit) for international shipments; implemented Odoo ERP and represented the company at the Chicago and Boston specialty coffee expos to build buyer and supplier relationships
Operations & Business Development
2020 to 2021 · Addis Ababa, Ethiopia
Operations and business development across a diversified Ethiopian industrial group covering real estate, coffee, furniture, and consumer goods.
- ▸Fully restructured the corporate vehicle fleet from 87 units down to 0, negotiating sales, loan compliance, and ownership transfers to eliminate carrying costs, loan obligations, and operational overhead tied to depreciating assets
- ▸Coordinated third-party and bank-appointed asset valuations for loan renewals, and audited prefinancing agreements and construction/manufacturing milestones before financing was released, ensuring proper documentation and regulatory compliance
- ▸Ran market and competitor analysis to source cost-effective procurement and negotiate favorable supplier contracts
Earlier Roles
Architecture Intern
Bereket Tesfaye Consulting Architects
March 2020 to June 2020 · Addis Ababa, Ethiopia
Made 3D models and hand-carved designs for luxury apartments and office buildings.
Assistant Project Manager
Nayna Rotaract
February 2019 to February 2020 · Addis Ababa, Ethiopia
Helped raise money to build a primary and middle school for an orphanage.
Education.
Where I studied and what I focused on
Master of Science
Finance
Hult International Business School
Cambridge, MA · 2025GPA: 3.58
Focus Areas
Key Coursework
Master of Science
Business Analytics
Hult International Business School
Cambridge, MA · 2026GPA: 3.79
Focus Areas
Key Coursework
Hult International Business School
Bachelor of Business Administration, Entrepreneurship (Minor: Finance)
Cambridge, MA
2024
Harvard University Extension School
Coursework, Finance, Computer Science & Security
Cambridge, MA
July 2023
Certificate in Quantitative Finance (CQF)
Certificate, Quantitative Finance
Online
In Progress
Skills & Tools.
What I work with day to day
Let's Connect.
If you want to talk about coffee, finance, data, or working together on something — reach out. I'm easy to get in touch with.