← All organizations

Financial data, analytics, and AI workflow software

FactSet

FactSet provides financial data, analytics and workflow software for asset managers, wealth managers, bankers and other investment professionals. Its FactSet Workstation, portfolio analytics, trading tools, data feeds and APIs support investment research, portfolio construction and trade execution. Developers can use curated datasets and APIs to build AI workflows, while its platform connects proprietary financial data with third-party sources and clients’ own datasets.

Founded in 1978 by Howard E. Wille and Charles J. Snyder, FactSet is now led by CEO Sanoke Viswanathan, who took the role in 2025. Its data architecture uses entity identifiers, symbology and concordance to align securities across asset classes with reference data and portfolio holdings. This connects otherwise disparate datasets for analysis, risk management and client reporting, both within and outside FactSet’s platform.

FactSet sells subscription services and reported more than 9,100 clients and 247,000 individual users in July 2026. Its completed 2025 acquisition of LiquidityBook, announced at a gross cash purchase price of $246.5 million, expanded its trading offering with order management and investment book of record capabilities. LiquidityBook’s integration with the Workstation links research and portfolio construction to order creation and execution, extending FactSet’s role across the investment workflow.

www.factset.com

2 talks

Newest first

1 speaker at AIE

Affiliations reflect their AIE appearances, not necessarily current employment.

Start here

  1. How to Build Planning Agents Without Losing Control - Yogendra Miraje, FactSet

    Start here to learn the distinction between static workflows and agentic systems, with an NVIDIA earnings-call preparation example grounding the architecture.

    Yogendra MirajeAI Engineer World's Fair 2025

  2. Skills are new features: Building Skill-Centric Harness — Yogendra Miraje, FactSet

    Continue here to learn how skill descriptions route financial-research and HTML/PDF report-generation tasks.

    Yogendra MirajeAI Engineer World's Fair 2026

Messages from the stage

Constrain planning through explicit structure

Miraje adapts LLMCompiler's plan-and-execute architecture into LangGraph nodes for blueprint generation, planning, execution, and joining results. Task blueprints limit planner complexity, while MCP-compatible tools supply explicit contracts.

Separate reusable skills from their execution harness

Miraje describes replacing proprietary blueprints with Anthropic's Agent Skills standard, combining selective skill loading with agentic loops. Human-reviewed admission controls govern which skills enter enterprise registries.

Affiliations reflect each recorded session, not necessarily current employment.

Company sources · checked 2026-08-28