Inside Asimov: BlackRock’s Platform Architecture for Applied AI at Scale

QCon San Francisco 2026

Session

Inside Asimov: BlackRock’s Platform Architecture for Applied AI at Scale

Tuesday Nov 17 / 11:45AM PST, Ballroom A at Hyatt Regency, San Francisco

Register

$2,955, Conference (3 days)
Current pricing ends October 13th

Speaker

Abstract

BlackRock’s Portfolio Management Group is transforming how investment teams conduct research by combining deep investment expertise with shared AI capabilities. Asimov, BlackRock’s AI-powered investment research platform, brings together data, investment context, and specialized AI agents to help investors research companies, monitor changing information, and generate insights at scale. Underneath Asimov is a layered architecture that separates data, intelligence and context, and applications and agents—allowing each to evolve independently while sharing enterprise integration, security, and controls.

The central design question is: how do you build an AI research system that scales across investment teams while preserving the flexibility for investors to develop workflows around how they actually research?

We will show how Asimov addresses this through two examples. Continuous thesis monitoring processes approximately 150,000 documents each week to keep investment context current across 10,000 companies. We will also show how the platform enables investors and domain experts to build their own AI-powered research experiences, with more than 1,000 dashboards and applications created and more than 30 promoted into a shared catalog for broader reuse.

Main Takeaways

  • Architect AI research around distinct data, intelligence/context, and application/agent layers, allowing capabilities to evolve and scale independently.

  • Turn large volumes of raw information into fresh, permission-aware investment context that can be reused across research workflows.

  • Combine shared AI capabilities with domain expertise, giving investors the flexibility to create workflows and agents around how they actually conduct research.

  • Turn successful user-built workflows into tested, reusable capabilities that can be discovered and adopted by other investment teams.

  • Create a standardized path from experimentation to production with access controls, evaluation, monitoring, and clear ownership—without centralizing every research workflow or decision.

Register

$2,955, Conference (3 days). Current pricing ends October 13th. All pass options.

76% senior dev or higher
1:11 speaker ratio
60+ practitioners

QCon San Francisco 2026 is a three day conference for senior software engineers, architects and team leads. An international program committee of working engineers selects every session. Patterns and practices, not products and pitches.

Share

LinkedIn X

Part of the track

Engineering AI Systems Hosted by Melanie Zhao Engineering Lead @BlackRock, Pioneering AI Adoption in Asset Management

From the same track

Tuesday 17 November

10:35 Ballroom A Session AI/ML Skills, Memory, or Fine-Tuning? The Engineering Loop Behind Self-Improving Agents Abhinav Sinha CEO @Lucidic AI, Previously @Stanford AI Lab, @Citadel and Susquehanna International Group, and @Apple 11:45 Ballroom A Session Inside Asimov: BlackRock’s Platform Architecture for Applied AI at Scale Dan Wolf, Sri Annapurna Bhupatiraju 13:35 Ballroom A Session AI The Revenge of the Data Scientist: Why Reliable AI Needs Evals, Traces, and Metrics Hamel Husain Machine Learning Engineer, 20+ Years in Applied AI, Machine Learning, and Data Science 14:45 Ballroom A Session AI/ML Lessons from Building a $100M Product in Six Weeks at OpenAI Brian Yang Member of Technical Staff @OpenAI 15:55 Ballroom A Session Inference Progressive Failure Modes of Modern AI Serving Systems 17:05 Seacliff D Unconference Unconference: Engineering AI Systems

Helpful links

Full schedule Conference homepage LLMs.md

Current pricing ends October 13th
$2,955, Conference (3 days)

Register