
Finance
|
Compliance Automation
|
YC W26
|
Valuation:
Undisclosed

Last Updated:
March 24, 2026

Builds autonomous AI agents that automate back-office compliance workflows at banks, including AML, KYC, KYB, sanctions screening, and SAR filing, with bulletproof audit logs.
AI agents for AML, KYC, KYB, sanctions/PEP monitoring, SAR filing. 10x analyst productivity, plug-and-play integration, no data migration. Compliance-first design with audit-ready documentation. SOC 2 compliance across all 5 Trust Services confirmed.
Michael M.'s Apple privacy-preserving ML background (used by billions of devices worldwide, invented techniques to train ML models without private data) hints at differential privacy and federated learning not yet publicly marketed. API-first overlay architecture targets compliance officers directly and integrates with existing bank stacks without system migration. Stealth customer pilots likely.
AI agents autonomously investigate AML and KYC alerts by gathering evidence, cross-referencing data sources, and drafting case narratives—reducing analyst workload by up to 10x.
An AI detective reviews suspicious bank transactions, pulls together all the evidence, and writes up the report so human analysts only need to approve it instead of spending hours doing it themselves.
It's like having a tireless junior analyst who reads every regulation, checks every database, and writes perfect case notes at 3 AM—except it never asks for coffee or puts in a transfer request.
AI agents automatically draft Suspicious Activity Reports (SARs) with complete narratives, evidence packages, and regulatory formatting—turning a multi-hour manual process into minutes.
An AI writes up the official suspicious activity paperwork for regulators, complete with all the evidence and proper formatting, so compliance teams just review and submit.
It's like having a legal secretary who instantly turns a detective's messy case notes into a perfectly formatted court filing—except it also double-checks every fact and never misspells "suspicious."
Leveraging the co-founder's Apple privacy-preserving ML expertise, Fenrock AI is positioned to enable banks to benefit from cross-institutional risk intelligence without exposing raw customer data—a novel approach to collaborative financial crime detection.
Banks can learn from each other's fraud patterns without ever seeing each other's customer data, like neighbors sharing crime alerts without handing over their security camera footage.
It's like a neighborhood watch where everyone's smart doorbell contributes to a shared crime heat map, but nobody can see inside anyone else's house—and the guy who built it literally designed the privacy system for a trillion-dollar tech company.
Charu Sharma previously founded one of the largest healthcare API companies (backed by General Catalyst's Hemant Taneja, scaled to 6M+ patients and 100+ employees) and a Future of Work/AI company with clients like Splunk, Twitter, and Coca-Cola. Michael M. built Apple's first privacy-preserving machine learning at scale, used by billions worldwide, and invented techniques to train ML models without private data, giving rare credibility for handling sensitive banking data with techniques competitors haven't operationalized for compliance.