Atlas Capital · 2025

Autonomous research agents

A fleet of investment‑research agents that read filings, run financial models and draft memos — supervised, not autonomous.

Tag
AI Agents
Duration
6 weeks
Role
Agent design · Eval · Frontend
Autonomous research agents
Memos / analyst / week
3.2× more
Time on bookkeeping
−68%
Hallucination rate
<0.4%
◇ The problem

Where they were stuck.

Junior analysts were spending 70% of their week on bookkeeping work — reading 10‑Ks, building DCFs, formatting memos.

◇ Our approach

How we shipped it.

  1. 01

    Decomposed analyst work into 9 deterministic + LLM-graded sub-tasks.

  2. 02

    Built a tool-using agent with strict schemas, caching and replayable traces.

  3. 03

    Designed a memo authoring surface that lets analysts edit any agent step.

  4. 04

    Hardened with an LLM-judge eval suite over 1,200 historical memos.

◇ Stack

What it's made of.

LangGraphPythonPostgresReactOpenAIAnthropic
Next case study

Cross‑platform clinician app

Lumen Health

Have a problem this hard?

Start a project →