Labs

The Problem

Small-cap IR teams face enterprise-level compliance requirements with no enterprise budget. Missing disclosures create real legal exposure: the average SEC securities settlement in 2024 was $42M, with a median of $14M. In October 2024, the SEC enforced against Unisys, Avaya, Check Point, and Mimecast for understating cyber risks they knew had "already materialized." 2026 is being called the "most complex reporting season yet." New cybersecurity 8-K rules, AI disclosure scrutiny, California climate requirements. The companies that get caught are the ones who didn't know what they were missing.

What I Built

Otlet is a peer intelligence platform for small-cap public companies ($50M-$500M market cap). Named after Paul Otlet, who envisioned a searchable global knowledge network in the 1930s. I analyzed 68 small-cap companies across biotech, software, and fintech. 82% had at least one disclosure gap. 46% had critical gaps (missing disclosures peers are making). The most common gaps: Climate/ESG (49% missing), Cybersecurity incident response (35%), AI/Automation risks (28%). Core features include Gap Detection ("What risk disclosures are my peers making that I'm not?"), 8-K Material Event Alerts (know when competitors file material changes), and Quarterly Filing Diffs ("What changed in peer 10-Qs this quarter?"). The difference from tools like Versance: they require you to know what to ask. Otlet proactively tells you what you should be paying attention to.

The Result

Otlet is currently in alpha testing. Phase 1 build includes gap detection on 10-K Item 1A (risk factors), 8-K material event alerts, and quarterly filing comparison. Target users are 1-2 person IR functions at small-cap companies: IR analysts, corporate secretaries, CFOs wearing the IR hat. Research shows 95% of IR professionals monitor competitors, but 59% spend less than $10K/year doing it manually. Interested in early access? Get in touch.

Tech

The Problem

Living in Canada, I noticed a recurring problem: finding a masjid meant searching Google for nearby mosques, finding their website to check prayer times, then figuring out how far away each one was. It was a proper task just to find salah by jamat. The existing solutions were products sold directly to masajid with their own apps, so while one app might have a couple of masajid near you, if a masjid hadn't onboarded with them, you were back to hunting through websites.

What I Built

After much experimentation (bots that scraped through various means, passing screenshots to AI, unpractical community platforms), I realized scraping wasn't the answer. Every masjid website is different, the smallest changes could break things, AI processing was expensive, and some masajid may not appreciate being scraped. So I designed a system that the community could maintain without scraping. Through trial and error, I built a UX that allows for easy updates and tracking. The system now allows easy finding and adding of masajid in Canada, US, UK, and Australia.

The Result

A product I'm taking into public beta to get feedback and continue building and expanding.

Tech