Torch Capital · Fund in New York Tech

Overview

Torch Capital is a venture capital firm investing in AI-powered companies that transform how consumers and businesses operate[1][3]. The firm has participated in funding rounds for AI-focused startups in the New York market. Torch Capital participated in the New York round for Baselayer's $20 million Series A funding in July 2026[2].

Leadership & org chart

JKJonathan KeidanFounder, Managing Partner

PP2 alumni

Investments

F2
F2 is the AI platform for private markets investors.

In the news

Reposted scott belsky@JKeidanX · Sep 27, 2026
Repostedscott belsky@scottbelskyWhile I am sure banks will work around and ultimately capitalize on agents representing consumers, this opens up the MANY larger questions around the elimination of "lazy margin" - the profit companies made out of our own laziness or negligence.
The Weekly Notable Startup Funding Report: 9/28/26The notable startup funding rounds for the week ending 9/26/26 featuring funding details for Hydrosat, Numeral, Firecral, and nineteen other deals representing $2.5B in new funding that you need to know about.AlleyWatch · Sep 27, 2026
Reposted Kahlil Lalji@JKeidanX · Sep 27, 2026
RepostedKahlil Lalji@bykahlilI am having all of these conversations, so let me tell you what is about to happen in banking over the next few years. There are roughly ~4,000 banks in the US, and the vast majority still run on legacy cores like FIS, Fiserv, and Jack Henry. These systems were not built for modern programmatic banking, let alone agentic payments, where the entire financial stack needs to be exposed through programmable surfaces. Agents are going to accelerate consolidation across banking. They will route activity toward financial institutions that are the most programmable, the easiest to integrate with, and offer the most rational economics. Unlike humans, agents do not care about branches, brand affinity, or decades-old banking relationships. That creates enormous pressure on both ends of the market. Across both G-SIBs and the long tail of regional and community banks, deposits, payment volume, and eventually credit activity will increasingly concentrate around a much smaller number of institutions built to support programmatic financial activity. The result will be significant compression in bank balance sheets and payment volume across the industry, accelerating consolidation and, in some cases, bank failures. The banking system was built around human distribution. Agents are going to reprice that distribution advantage very quickly.
Reposted Method Security@JKeidanX · Sep 10, 2026
RepostedMethod Security@method_securityMethod and Palantir are announcing the Cardinal Program: a consortium of AI, infrastructure, and security partners to strengthen national cyber resilience. America’s municipalities, water and power utilities, and critical infrastructure operators now face cyber adversaries armed with AI. Their defenders should not have to stand alone. Cardinal will provide select municipalities, utilities, and critical infrastructure operators with free, opt-in, continuous autonomous security assessments - with expert operators and human oversight. If your organization protects critical systems or public services, we want to hear from you.
Reposted a16z@JKeidanX · Sep 9, 2026
Reposteda16z@a16zFrom @patrickc, examples of people quickly accomplishing ambitious things together: https://t.co/qNGyan0Icu
Reposted Alyson Shontell@JKeidanX · Aug 26, 2026
RepostedAlyson Shontell@ajsSome of the earliest money in Cursor (the largest vc-backed acquisition ever) turned a $750,000 investment into a ~$1 billion return. @davidtisch and his midas touch at Box Group strike again! Great profile here of one of the best early stage investors who wins while going against the grain @agarfinks https://t.co/hirdBa5RnP
Reposted Jesse Middleton@JKeidanX · Aug 20, 2026
RepostedJesse Middleton@srcasmBefore a founder has traction, I'm looking for one thing: clock rate. How fast do they learn? How fast do they adapt? When a new frontier model drops, do they say "holy shit, that means I can do so much more" or do they start worrying about what it replaces? The founders playing offense against foundation models are the ones I want to back. The ones playing defense are usually building something that's about to get eaten. We saw this with Arcee. When they started, their bet was that large enterprises would want small, specialized models. People thought they were early. They were. But they were right on direction. Today, open weight models, model merging, fine-tuning, RL, and new hardware make enterprise SLMs accessible without a $100M+ spend. The market caught up to the bet. That's the kind of founder clock rate I'm looking for at @flybridge. If you're building and your instinct when a new model drops is to go on offense, my DMs are open.
RT @sourceryy: Founding Architect of the Thiel Fellowship @DStrachman shares the top 3 traits they look for: 1. Hyper Fluency: "They have@JKeidanX · Aug 11, 2026

Something wrong or missing? Send an update. Fixed within 24 hours.