Finance on Lynx Tech Blog

Regulatory Radar Weekly Report 2026-09-21 ~ 2026-09-28: Summary of China's Financial Regulatory Enforcement Actions

Mon, 28 Sep 2026 00:00:00 +0800

This Week at a Glance This week (2026-09-21 through 2026-09-28), five regulatory channels published a total of 49 penalty and disciplinary notices: NFRA 32, AMAC 11, SSE 6. As of this issue, the LynxRegulus regulatory radar has accumulated 448 records, of which 245 have completed full-text structured parsing (penalty amounts and involved parties). This Week’s Details Date Source Type Title Link 2026-09-24 SSE Regulatory Work Letter Regulatory Work Letter on Share Price Volatility Matters R

Retiring ERPNext (6 Containers) — Running Odoo 18 Chinese Accounting on Port 8090

Thu, 24 Sep 2026 21:30:00 +0800

Bottom line first: port 8090 is still port 9090, but the tenant inside has changed. The original six-container ERPNext v15 stack has been decommissioned entirely — compose stopped, volumes deleted, all images pruned, the instance directory removed, and the corresponding hosts entry wiped. The port now hosts Odoo 18 Community + PostgreSQL. A Chinese database was created with the small-business accounting standards chart of accounts and 13%/9%/6%/3% VAT rates pre-configured. The first test expens

Before Firing My Bookkeeping Agency, I Sent 134 AI Agents Down the Rabbit Hole

Thu, 24 Sep 2026 18:30:00 +0800

Conclusion first: not a 100% replacement — but closer than you’d think. Open source plus AI can take over 80-90% of what a bookkeeping agency does (ledgering, reconciliation, invoice collection and parsing, draft vouchers). But there is one loop no software can ever close: tax filing submission and compliance liability. For small-scale taxpayers, end-to-end replacement is roughly 60-75% at a steady-state cost of 2-4 hours a month. One sentence: the ledger can be open source; filing cannot

What the Financial Industry Actually Tracks, What's Still Done by Humans, and Where Individuals Can Break In: An In-Depth Study of 114 AI Agents

Wed, 16 Sep 2026 13:39:00 +0800

I previously ran a round of research and concluded that finance is one of the industries with the deepest AI penetration. Deep penetration means a solid data foundation, many players, and potentially more opportunities. This time I broke the question into three parts and ran them through a multi-agent deep-research pipeline for 97 minutes: What does the financial industry need to track in real time? Which tracking and processing tasks are still manual repetitive labor? Where can individual devel