AlphaScreen

Where it started

AlphaScreen started from a workflow problem rather than a model problem. Good equity research often breaks down because the process is scattered across too many places: one tool for screening, another for notes, another for comparison, and then a separate layer of interpretation on top.

That fragmentation slows everything down and makes conviction harder to build. I wanted to explore what it would look like to bring those steps into a single, more focused research environment.

What needed to change

The pain points were fairly clear:

• analysts were losing momentum every time they jumped between tools,
• the reasoning behind a decision was easy to lose,
• and repeated evaluation workflows were taking more effort than they should.

The product direction

AlphaScreen reframes research as a guided workflow. It gives users a place to screen opportunities, investigate them with AI-assisted context, compare candidates consistently, and keep a record of why a thesis is starting to strengthen or weaken.

The goal was not to automate judgment away. It was to make the path to judgment cleaner and more deliberate.

What it is trying to improve

• A more focused workspace for ranking and filtering global equities.
• Faster context gathering when exploring a new idea.
• Better consistency when comparing several candidates side by side.
• A clearer chain of reasoning from first screen to final conviction.

Why it matters to me

I like products that reduce noise rather than add more of it. AlphaScreen fits that pattern. The value is not in giving the user endless output. It is in helping them stay close to the evidence, move more quickly, and explain their reasoning more clearly.

Because the project is private, I am intentionally keeping the technical details light here. The more interesting part, at least for this portfolio, is the product thinking behind it: better tools create better thinking environments.