Blog
Insights, analysis, and updates from the FinCatch team.
Recurring Monitoring: From AI Assistant to Proactive Research Partner
Recurring monitoring turns FinCatch from an on-demand assistant into a proactive research partner that keeps watching the market, judging relevance, and extending the user's research thread over time.
How FinCatch Is Different
How FinCatch uses a three-layered 'financial mind' to uncover second-order effects and structural market truths.
Agents, Personal Guidance, and Memory: How We Think About Building FinCatch
Why FinCatch is being built as a layered agent system that combines a market world model, personal memory, and active guidance for equity research workflows.
How We Built an Agentic Equity Research Pipeline
A behind-the-scenes look at how we redesigned equity research into a modular, agent-orchestrated pipeline that produces industry-grade reports with tighter runtime and better control.
How FinCatch Remembers What You Think About the Market
Most AI assistants forget everything the moment you close the chat. FinCatch doesn't. We built a structured knowledge graph that remembers your investment views, tracks how they evolve over time, and reasons across companies and sectors — so your AI actually knows you as an investor.
Why the Hard Part Isn't the LLM: The Application of Harness Engineering in FinCatch
Reliability in financial AI doesn't come from the model. It comes from the scaffolding. Here's how we built it.
Scenarios: How Our Research Agent Thinks About What Could Happen
Our agent has a step we call **simulation**, and it only runs when the question asks for it. A backward-looking question gets the facts and stops. Anything that touches outlook, exposure, or risk triggers it.
From Facts to Foresight: Rethinking the Knowledge Graph for Financial Markets
This post is about why we built our own financial knowledge graph, what makes it different, and how it changes the work of financial research.
Claw on Cloud: How We Run Stateful AI Agents on a Stateless Edge
How FinCatch keeps agent context warm across requests using a tiered container pool, R2-backed state, and skills as plain markdown.
Management Guidance Extraction and Analysis
How we extract forward-looking guidance from earnings transcripts and press releases, normalize it into a structured time series, and grade management against their own promises.
Building an Agentic Financial Modelling Framework
How we built a Python-based DCF modelling system designed for LLM agents, with reliability layers, sanity checks, and a UX that serves both agents and humans.
A Collective Unconscious for Markets: The Philosophy Behind FinCatch
How FinCatch uses a three-layered 'financial mind' to uncover second-order effects and structural market truths.
Beyond the Black Box: Explainable AI in Finance
In the high-stakes world of finance, "trust me, I’m an AI" is not a viable strategy. A single hallucinated statistic or anachronistic data point can trigger million-dollar losses.
Brief Breakdown of nano-graphrag: A Lightweight Alternative to GraphRAG
An exploration of nano-graphrag, a more hackable implementation of GraphRAG for knowledge graph-enhanced retrieval augmented generation