Investment research · Portfolio construction · Evidence-linked tools
Northeastern MBA and MS in Finance graduate, with institutional experience at AEW Capital Management, Loomis Sayles, Wellington Management and Prime Buchholz.
I focus on questions that matter to an investment decision: what changed, whether a signal adds information, and what a portfolio trade-off is actually worth.
Explore my interactive research portfolio
How much tracking risk is worth accepting to accommodate restrictions and reduce trading?
A historical portfolio-implementation study using WRDS/CRSP ETF returns, lagged risk estimates, drifting holdings and modeled transaction costs. In the no-gold case, about 60% less turnover saved only 1.10 basis points annually while adding about 27 basis points of tracking error. The useful finding is the trade-off, not a headline return.
Interactive case study · Methods and limitations
The research did not establish incremental return-predictive value. That negative finding informed an evidence-linked analyst workflow for examining disclosure changes and rejecting unsupported narratives.
The public prototype contains 995 evidence-backed disclosure changes. In eight bounded research-idea cases, two hypotheses survived, six were rejected and none produced an actionable trade view. It is a frozen prototype; formal human validation remains pending.
Explore the demo · Research findings · Product repository
A Black-Litterman research prototype combining analyst views, covariance estimation, constrained optimization and attribution for student-fund research. Implementation evidence is distinct from verified live use or improved investment performance.
A browser-based educational simulator for practicing market-making, options, execution and decision-making with synthetic scenarios. Public UAT preview; not a live-trading system or validated hiring assessment.
Try the simulator · Repository
- Equity-Factor-Model: factor research, portfolio construction and modeled implementation diagnostics. Results belong to their documented specifications, not a universal performance claim.
- Corporate Niche Dynamics: measuring competitive crowding in business-model space. Synthetic benchmark only; no real-company investment claim.
- Optlab Research: research-platform architecture and portfolio-analysis infrastructure.
- Earnings reactions: evaluating whether structured AI analysis adds explanatory value under controlled information cutoffs.
- Liquidity forecasting: studying whether volume forecasts improve modeled implementation at a controlled tracking-risk level.
Both are in development. Source and engineering audits are underway; no empirical performance is claimed.
- Start with the economic question and a credible comparison.
- Preserve source provenance and point-in-time information boundaries.
- Separate historical results, modeled costs, prototypes and synthetic demonstrations.
- Keep negative findings and limitations visible.
These projects use AI-assisted research and development, with documented methods, source provenance and explicit limits on what the evidence supports.
MBA and MS in Finance, Northeastern University. BA, Hamilton College. Former Head of Equity Research for Northeastern's student-managed 360 Huntington Fund, January–May 2026.