Equity Research & Chart Analysis for Investment Desks
Capital Markets / Equity ResearchReads earnings transcripts, news and live charts across 500+ U.S. equities, assembling bull and bear cases for an analyst to rule on.
Overview
An in-house equity research desk covering more than 500 U.S. public companies needed thesis-ready analysis in minutes rather than hours per name. Canvas Labs Partners built a multi-agent research platform that reads earnings transcripts, news and live charts, assembles structured bull and bear cases, and leaves the final judgment call with the analyst.
500+ equities under coverage
bull and bear cases generated automatically for every name
60% less idea-generation time
the analyst still makes the final judgment call
Two hours to thirty minutes
full chart and technical analysis turnaround
Challenge
Forming a thesis on a single name took an analyst over two hours of reading earnings transcripts, news and charts. Across a coverage universe of more than 500 U.S. public companies, that reading time set the pace of the desk's whole idea-generation cycle. The firm wanted the turnaround in minutes, without handing the final judgment call to a machine.
Solution
Canvas Labs Partners built a multi-agent pipeline that ingests earnings transcripts and news through custom function-calling tools. It runs across multiple LLM providers rather than one, with OpenAI, Anthropic and Vertex AI all sitting behind it. From that reading the pipeline auto-generates a structured bull and bear case for each name, and surfaces sentiment, breadth and sector-rotation signals across the coverage universe.
Charts are handled by a separate multimodal agent that works in vision and text together. Playwright drives a headless browser to screen-capture live charts, which the agent reads as images, working through async multi-step reasoning to produce the full chart and technical read that previously took an analyst more than two hours.
A later extension turns finished research into published content. An agentic content pipeline orchestrates OpenAI, Anthropic and NotebookLM, with FFmpeg handling the audio processing, and formats a completed piece of research for multiple social platforms, then publishes it automatically rather than waiting on someone to repackage it by hand.
Results & benefits
More than 500 U.S. public companies now sit under automated coverage, each with a structured bull and bear case generated automatically. Idea generation takes the desk 60% less analyst time than it did, and a full chart and technical read that ran beyond two hours now completes in about thirty minutes. Turning a finished piece of research into published social content dropped from hours to roughly five minutes. The analyst still makes the final judgment call; what changed is how much reading has to happen first.
Customer
In-house equity research desk
Services
- Agentic AI consulting & development
- Multimodal AI engineering
- Research & content workflow automation
Tech stack
About us
Canvas Labs Partners is a software consulting team that designs and ships production agentic AI systems: multi-agent pipelines, tool-calling architectures, and human-governed automation, together with the trading infrastructure, blockchain, mobile, and product engineering those agents plug into. We publish our work as a case ledger rather than a portfolio, so every claim on this site links to the engagement behind it. Engagements follow the same four-step discipline, Map, Bound, Instrument and Govern, whatever the stack, and every entry in the ledger carries a result we are willing to be measured against.
Delivered individually, in a prior role, by one of Canvas Labs Partners’ founding team members. Not by the firm itself.
Have a system like this one?
We built this one end to end. If you are facing something similar in capital markets, book a free discovery call and we will tell you what it would take.