Eddy R. Lora

Eddy reading the back of a book between the shelves of a bookstore, in black and white.

Eddy at night beside a lamp post in a park, in black and white.

Eddy lining up a shot on a pool table in a bar, in black and white, a friend waiting with a cue.

About.

I do equity research, and I build the tools I use to do it. I love building and working on complex projects, and I keep taking on the ones that let me express myself technically.

I write initiation reports and the models behind them, mostly on industrials and tech hardware. I also run Reminous, a small agency where I build creative and acquisition systems and run the ads.

Before Rutgers I went to Woodberry Forest in Virginia, where I played football and basketball, ran track and was head reader in chapel. At Rutgers I studied English with a minor in history and made the Dean's List.

Woodberry is where I started to find what I liked, such as debate and literature. I didn't realize it until I graduated, but in a tight-knit community I was part of a lot of conversations that didn't seem like much at the time and kept challenging what I believed. Through hard conversations, rejection and other setbacks, I got comfortable with failing as part of getting better. The English program at Rutgers worked the same way. The classes were small, everyone took part in the discussion, and we usually left still working through it.

In 2023 I interned on two equity research desks at the same time while in school full time. At TD Bank I finished first in my region in my first full quarter. I also co-founded a marketing software company, Magnifye, and carried the business forward under my own agency, Reminous. I passed the SIE in September 2026, on my second attempt.

Away from work I enjoy chess, basketball, golf, weightlifting and building PCs.

Rutgers, The State University of New JerseyBachelor of Arts in English, minor in History. Dean's List. May 2025.

Woodberry Forest School, VirginiaFootball, basketball, track. St. Andrew's Chapel head reader. May 2021.

I've published one initiation, on Modine Manufacturing, and built two more, on TE Connectivity and Eaton. The other tickers shown are names my trading engine scans.

Modine Manufacturing, an initiation.

July 27, 2026, 34 pages

A 34-page initiation on Modine Manufacturing, a company rapidly shifting its product line toward data center cooling. I published it two days before the company reported its first-quarter earnings.

Read the note

Page three of the Modine note, the key takeaways.

03

What	Old	New	Why
Corporate cost in my bull and bear cases	$88.4 million, the GAAP figure, next to adjusted numbers	$62.6 million, on the adjusted basis Modine re-issued July 24	Keeps every case on the same basis.
The margin the rule watches	Calculated by hand each quarter	A model column that calculates it, 15.29 percent for the quarter	Removes manual error.
Data Centers growth for the quarter	A growth rate mixing two sets of segment figures	Removed, with a note in the cell	The two sets of figures weren't comparable.
The full-year check for fiscal 2027	Ran against a single quarter	Turned off, with a note in the cell	One quarter isn't enough to check a full year.

Four changes I made to the model after the first quarter.

My model.

Every historical number in my models traces to a filing, and every forecast number to a dated assumption. Before a note goes out, every number in it is checked against the model.

After the first quarter, I rolled the model forward and logged every change. All nine reported lines matched the company's figures within $0.05 million.

The margin the rule watches fell under 16.0 percent in the first quarter after the note.Source: Modine's first-quarter fiscal 2027 release, July 29, 2026, and the segment history Modine re-issued on July 24, 2026.

The rule in the note.

Before publishing, I wrote a rule into the note. If the combined adjusted EBITDA margin of Data Centers and Commercial HVAC falls below 16.0 percent for two quarters in a row, I lower my rating.

It came in at 15.29 percent for the first quarter, the first reading below 16.0 percent in the segment history Modine re-issued on July 24. The rule requires two quarters in a row, so the rating stays until the second quarter reports.

Python and Claude.

I model in Excel, by hand. I use Python to pull filings, and Claude Code to help read them, check my first pass at the model and review the written note.

I pull ten years of filings.
I build the model from them in Excel.
Every figure is checked against its filing before anything is published.
Claude reviews the draft and flags where an analyst would push back.
I review all of it, decide the rating and write the thesis.

The prompt, the run, and the three notes so far.

My sell-side research experience.

Two internships in 2023, held at the same time while I was a full-time student at Rutgers.

C.L. King & Associates Hoboken, New Jersey. May to September 2023.

This is where I decided on equity research. I worked under a senior analyst with thirty years of experience, and for part of the summer I was his only support.

I maintained his models on specialty chemicals and materials companies. He then had me build the model and prepare the initiation on Rogers Corporation, a company he hadn't covered before. That meant the industry and peer work, the historical model and forecast, calls with management and investor relations, and the morning call. It's the same process I used on Modine, Eaton and TE Connectivity.

Auerbach Grayson New York. April to October 2023.

I worked the early shift. Research from partner brokers in Asia and Europe came in overnight, and before the US open I read each report, summarized the key points by ticker, confirmed them with the analyst and sent them out.

Securities Industry Essentials (SIE), passed September 2026.

My trading experience.

I started trading in August 2022, during the sell-off in growth stocks. I traded semiconductor names and the S&P through the decline and the recovery, mostly short-term trades at the open on a fifteen-minute chart, with the one-minute for entries. In 2024 I closed 128 trades, most of them in Nvidia and SPY.

I kept my charts simple, with trend lines, supply and demand zones, volume, the 50-day moving average, and the 9-period moving average on shorter timeframes. I entered on a candle close, not on a touch of the level.

I logged every trade in a journal of nineteen fields, with entry and exit times to the second and the reason for each exit. My rules were three to one reward to risk, the exit set before the entry, and every trade that met the setup taken, none that didn't. In October 2024 I started grading each setup, from a basic setup at a level up to one with volume and a dated catalyst, and sized positions by grade. I didn't test whether the grades worked until 2026.

Source: daily bars read from my trading engine, November 2023 to July 2024. The daily bars are real prices, and the movement inside each bar is simulated for the animation.

Trade #
Date
Ticker
Entry Price
Exit Price
Strategy
Exit Signal
Strikes
Type
Time of Day
Entry Time
Exit Time
Duration (minutes)
Entry Mark
Exit Mark
Difference
three columns withheld

Source: the header of my own journal, September 2024, three columns withheld. The trades themselves are not published.

Testing my own rules.

In 2026 I rebuilt those rules in Python, so the same code grades setups in the backtest and in the live scanner.

Price data for every name, pulled and cleaned.
One set of rules grades every setup, historical and live.
1,771 trades replayed over a 156-name universe, 2010 to 2026.
The grades tested against 20,000 permutation and 20,000 block bootstrap draws.
The grades separated in the years I built on, but not in the years held out, so I don't size trades on them.

364tests the code runs, every one passing
156names the scanner reads
1,771trades replayed, 2010 to 2026
20,000permutation draws
20,000block bootstrap draws

Read from the engine when this page was built. If one test fails, the page does not build.

pytest

$ python -m pytest........................ [ 19%]........................ [ 39%]........................ [ 59%]........................ [ 79%]........................ [ 98%].... [100%]364 passed in 114.21s (0:01:54)

The engine's last test run, read when this page was built.

The line is the actual gap in R, return as a multiple of the risk taken, between B+ and C setups, the two grades with enough trades to measure, against ten thousand random shuffles of the grades in each period. In the years I built on, it falls outside almost every shuffle. In the held-out years, it falls inside them.Source: the engine's own null run, 20,000 permutation and 20,000 block bootstrap draws, re-run read-only when this page was built and matched to its saved summary.

Reminous.

In October 2024 I co-founded Magnifye, marketing software for restaurants, and when owners asked us to run their ads, we started doing that too. My co-founder left, we dissolved the company, and I continued the work under my own company, Reminous.

Reminous runs ads for business owners and builds everything after the click, including landing pages, lead systems, follow-up and weekly reporting.

It's also where I learned to build automated systems, which I now use in my research.

reminous.com

The whole Reminous site, top to bottom, scrolling by itself in a browser window.

How it runs, on the Reminous site: the calls the team makes by hand, and the reading AI does for them, each step marked done.

The weekly report on the Reminous site, a dashboard for a made-up week of ads for a made-up business.

The Reminous site, live at reminous.com. The client work on it is real. The example business is a sample and is labeled as one.

The systems behind it.

Each of these is a live workflow, pulled from n8n, the automation tool that runs them.

n8n Lead Bus

Drag the drawing sideways to follow it.

Lead Bus. Takes leads from the site form, standardizes them, filters spam and logs them. The office is alerted right away in business hours, or at nine the next business morning, and the lead gets a text only if they consented.Meta Lead Watcher. Reads leads from the ad form as they arrive and sends them through the same alerts and consent check.Boost Form Lead Bridge. A boosted post created its own lead form with nothing connected to it. This pulls those leads from the ad platform into the same sheet.Inbound SMS + Voice. Calls to the business number ring through. Text replies are verified, emailed to the office at any hour and texted on during business hours, and calls are logged.Lead Heartbeat. Every three hours, checks recent ad leads against the alerts that went out and alerts me if any were missed.Error Handler. When a workflow fails, it checks whether the failure is real, then alerts me by email and text.Spend Watchdog. Checks daily and monthly ad spend against the limit every hour and alerts me if either goes over.Weekly Report. Every Monday morning, compares the week's ad results to the week before, with the lead log, and sends the report.Uptime Watch. Checks the site's main pages and n8n itself on a schedule and alerts me if anything is down.SMS Delivery Watch. Catches every failed text, from the carrier's callback and a sweep of recent messages, and reports it.Twilio Balance Watch. Each morning, estimates how long the texting balance will last and alerts me by text and email when it runs low.Twilio Usage Alert. Turns the carrier's usage alerts into a readable email.Consent Trail Export. On the first of each month, exports the consent records to a file and emails it as a backup.n8n Self-Backup. Every night, exports every workflow to a file and emails it as a backup.

My App.

Coming soon

A training and competition app for Excel, Python and AI skills, built like chess.com for finance. You work in a real spreadsheet or terminal, get graded on your keystrokes, race your best run or the daily challenge, and review your game afterward. The spreadsheet engine has 103 automated checks behind it.

It's free and doesn't need an account. I built it to practice the same desk skills until they're automatic.