正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Learn how J.P. Morgan developed a multi-agent system architecture for investment research. David Odomirok and Zheng Xue reveal how they built "Ask David" - a sophisticated multi-agent AI system designed to automate investment research for thousands of financial products. Watch the full session here:

LangChain

200,215 subscribers

32,327 次观看 • 1 年前 •via X (Twitter)

财经科学技术

Anya Rossi• Live Now

Private livecam show

5 条评论

Agent Blueprint 的头像

Agent Blueprint1 年前

@jpmorgan JPMorgan’s “Ask David” isn’t just a prototype. In 2025, it drove a 3.4x productivity leap, handled 28M trades/day, and redefined how AI scales across real capital markets. $18B+ invested, AI is now core infra.

Alger 的头像

Alger1 年前

Looking to invest in the Enablers and Adopters of AI? Consider an actively managed fund investing in companies actively involved in developing and implementing AI technologies.

Vishal 的头像

Vishal1 年前

@jpmorgan Great example of how multi-agent systems can handle complex, high-stakes tasks like investment research.

snehil prakash 的头像

snehil prakash1 年前

Fascinating insight into @jpmorgan's 'Ask David' system! The use of a multi-agent architecture for investment research is a game-changer. It’s impressive how it aims to automate and enhance efficiency while handling high-stakes financial data. The integration of LangChain's capabilities here is particularly intriguing. Looking forward to seeing how this scales and impacts the broader financial sector.

Josh 的头像

Josh1 年前

@jpmorgan And what they aren't on x haha

相关视频

JP Morgan's investment research team just shared exactly how they built their multi-agent system "Ask David", and it's the same architecture pattern showing up everywhere: - supervisor agent orchestrates - specialized subagents handle retrieval, structured data, analytics - LLM-as-judge reflection node before the answer ships - human-in-the-loop for the last accuracy gap worth watching for anyone building:

JP Morgan's investment research team just shared exactly how they built their multi-agent system "Ask David", and it's the same architecture pattern showing up everywhere: - supervisor agent orchestrates - specialized subagents handle retrieval, structured data, analytics - LLM-as-judge reflection node before the answer ships - human-in-the-loop for the last accuracy gap worth watching for anyone building:

Adam Ghowiba

1,979,086 次观看 • 2 个月前

🔥 Our latest LangChain Academy course – Deep Research with LangGraph – is now live! 🔥 Deep research agents are taking off – from major AI labs to companies building their own. Research is inherently open-ended. You can't always predict whether a question needs broad exploration or deep analysis. Agents excel here because they adapt on the fly, using each finding to decide where to dig next. Building these systems ourselves and with customers, we've learned that structure matters. The best research agents scope problems with users first. Then, they coordinate multiple specialists instead of overwhelming one generalist. LangGraph is built for these types of long-running, multi-agent workflows. Its persistence layer tracks progress across agents. LangSmith gives you the observability and evaluation tools you need to track and improve performance. In this course, you'll build and evaluate: - A user scoping agent to define research parameters - A multi-agent research team with supervisor - And add tool integration via MCP Enroll for free ➡️

🔥 Our latest LangChain Academy course – Deep Research with LangGraph – is now live! 🔥 Deep research agents are taking off – from major AI labs to companies building their own. Research is inherently open-ended. You can't always predict whether a question needs broad exploration or deep analysis. Agents excel here because they adapt on the fly, using each finding to decide where to dig next. Building these systems ourselves and with customers, we've learned that structure matters. The best research agents scope problems with users first. Then, they coordinate multiple specialists instead of overwhelming one generalist. LangGraph is built for these types of long-running, multi-agent workflows. Its persistence layer tracks progress across agents. LangSmith gives you the observability and evaluation tools you need to track and improve performance. In this course, you'll build and evaluate: - A user scoping agent to define research parameters - A multi-agent research team with supervisor - And add tool integration via MCP Enroll for free ➡️

LangChain

61,205 次观看 • 10 个月前

2️⃣ Hypothesis Generation Built with Co-Scientist, this system can help brainstorm and evaluate novel research ideas for open challenges. Using a multi-agent “idea tournament”, the system generates, debates and evaluates hypotheses to show what works, what doesn't, and why.

2️⃣ Hypothesis Generation Built with Co-Scientist, this system can help brainstorm and evaluate novel research ideas for open challenges. Using a multi-agent “idea tournament”, the system generates, debates and evaluates hypotheses to show what works, what doesn't, and why.

Google DeepMind

23,263 次观看 • 1 个月前

JP Morgan revealed the exact architecture behind "Ask David" their investment research AI agent. Supervisor. Subagents. Judge. Human. Here's the whole stack. → 1 supervisor agent routes and orchestrates → Specialized subagents handle retrieval, structured data, and analytics separately → LLM-as-judge reflection runs before any answer leaves the system → Human-in-the-loop patches the last mile accuracy gap AI can't solve alone Every serious agent team is converging on this pattern. JP Morgan just confirmed it works at enterprise scale. Build accordingly. worth watching 👇

JP Morgan revealed the exact architecture behind "Ask David" their investment research AI agent. Supervisor. Subagents. Judge. Human. Here's the whole stack. → 1 supervisor agent routes and orchestrates → Specialized subagents handle retrieval, structured data, and analytics separately → LLM-as-judge reflection runs before any answer leaves the system → Human-in-the-loop patches the last mile accuracy gap AI can't solve alone Every serious agent team is converging on this pattern. JP Morgan just confirmed it works at enterprise scale. Build accordingly. worth watching 👇

Priyanka Vergadia

17,075 次观看 • 2 个月前

Trade research drudgery for research breakthroughs — @SciSpace_ Agent gives back an average of 1300 hours a year to researchers. SciSpace Agent is the first #AI built exclusively for the scientific research community. Learn more about your new AI co-scientist in this thread.

Trade research drudgery for research breakthroughs — @SciSpace_ Agent gives back an average of 1300 hours a year to researchers. SciSpace Agent is the first #AI built exclusively for the scientific research community. Learn more about your new AI co-scientist in this thread.

Kirk Borne

11,871 次观看 • 11 个月前

I built a tool to fix Openclaw agent workspaces Fix and improve how your agent thinks and operates in 5 minutes Single or multi-agent setups, and based on real AI research + my last 6 weeks of non-stop Openclaw use Free Souls CLI + launching this week

I built a tool to fix Openclaw agent workspaces Fix and improve how your agent thinks and operates in 5 minutes Single or multi-agent setups, and based on real AI research + my last 6 weeks of non-stop Openclaw use Free Souls CLI + launching this week

toli

146,213 次观看 • 4 个月前

Introducing Mesa, the world’s first multi-agent code review system that truly understands your system and architecture to deliver senior-level reviews.

Introducing Mesa, the world’s first multi-agent code review system that truly understands your system and architecture to deliver senior-level reviews.

Oliver

65,152 次观看 • 8 个月前

Alan Waxman is the CEO of Sixth Street, one of the world's largest multi-strategy private capital investment firms. He spoke to Chairman and CEO David Solomon about how sports have become a core strategic investment theme for his business.

Alan Waxman is the CEO of Sixth Street, one of the world's largest multi-strategy private capital investment firms. He spoke to Chairman and CEO David Solomon about how sports have become a core strategic investment theme for his business.

Goldman Sachs

19,102 次观看 • 8 个月前

We believe AI can be a dedicated research partner to help discover the next breakthrough. Enter Co-Scientist: our latest Gemini-based multi-agent system that can generate, debate and evolve novel hypotheses for complex scientific problems 🧵

We believe AI can be a dedicated research partner to help discover the next breakthrough. Enter Co-Scientist: our latest Gemini-based multi-agent system that can generate, debate and evolve novel hypotheses for complex scientific problems 🧵

Google DeepMind

196,828 次观看 • 1 个月前

⚕️ Multi-Agent Medical AI Assistant A full-stack system that combines agents, computer vision, hybrid RAG, and voice I/O into a unified pipeline for diagnosis, research, and patient interaction. The architecture includes: • LangGraph by LangChain for agent orchestration • PyTorch vision models • Hybrid RAG on Qdrant + Hugging Face re-ranking • Unstructured for parsing medical PDFs • tavily for real-time search • ElevenLabs voice interface • FastAPI + Docker backend • Human-in-the-loop validation Built by Souvik Majumder 🚀 Code:

⚕️ Multi-Agent Medical AI Assistant A full-stack system that combines agents, computer vision, hybrid RAG, and voice I/O into a unified pipeline for diagnosis, research, and patient interaction. The architecture includes: • LangGraph by LangChain for agent orchestration • PyTorch vision models • Hybrid RAG on Qdrant + Hugging Face re-ranking • Unstructured for parsing medical PDFs • tavily for real-time search • ElevenLabs voice interface • FastAPI + Docker backend • Human-in-the-loop validation Built by Souvik Majumder 🚀 Code:

Qdrant

34,045 次观看 • 1 年前

I built a multi-agent Deep Research team using Qwen 3 that can generate consulting style reports as Google docs. It has 3 AI Agents working together as a team: 1. Research Question Agent 2. Web Search Agent 3. Google Docs Compilation Agent 100% Opensource Code.

I built a multi-agent Deep Research team using Qwen 3 that can generate consulting style reports as Google docs. It has 3 AI Agents working together as a team: 1. Research Question Agent 2. Web Search Agent 3. Google Docs Compilation Agent 100% Opensource Code.

Shubham Saboo

25,071 次观看 • 1 年前

💫 Simplify your AI agent development. Google's new open-source Agent Development Kit (ADK) is designed for effortless multi-modal and multi-agent systems. Get started →

💫 Simplify your AI agent development. Google's new open-source Agent Development Kit (ADK) is designed for effortless multi-modal and multi-agent systems. Get started →

Google for Developers

13,232 次观看 • 1 年前

Insights Agent & Multi-turn Evals Agents run for a long period of time and have multiple interactions with users, tackling a wide variety of problems. Today we’re launching two new features in LangSmith, our agent engineering platform, so you can better understand your agent behavior: • Insights Agent is a new agent that automatically categorizes agent behavior patterns in LangSmith • Multi-turn Evals lets you evaluate agent trajectory across complete conversations, so you can see if the agent actually accomplished user goals Learn more:

Insights Agent & Multi-turn Evals Agents run for a long period of time and have multiple interactions with users, tackling a wide variety of problems. Today we’re launching two new features in LangSmith, our agent engineering platform, so you can better understand your agent behavior: • Insights Agent is a new agent that automatically categorizes agent behavior patterns in LangSmith • Multi-turn Evals lets you evaluate agent trajectory across complete conversations, so you can see if the agent actually accomplished user goals Learn more:

LangChain

13,322 次观看 • 8 个月前

What if AI agents could hunt for yield on the web? We built a small demo using the TinyFish Web Agent API The agent navigates websites, executes multi-step research workflows, completely autonomously. Exciting for the future of autonomous yield strategies. #TinyFishAccelerator #BuildInPublic

What if AI agents could hunt for yield on the web? We built a small demo using the TinyFish Web Agent API The agent navigates websites, executes multi-step research workflows, completely autonomously. Exciting for the future of autonomous yield strategies. #TinyFishAccelerator #BuildInPublic

YieldFi

10,901 次观看 • 4 个月前

Preview of our DeepResearchSwarm™ It's incomprehensibly fast as it leverages a multi-agent architecture instead of just one agent. It utilizes a joint hierarchical-parallel architecture that creates a plan and then concurrently executes the plan with multiple search agents. Workflow: Task → Research Director → multiple search agents → final summary by summarization agent. Stay tuned for the release ↓

Preview of our DeepResearchSwarm™ It's incomprehensibly fast as it leverages a multi-agent architecture instead of just one agent. It utilizes a joint hierarchical-parallel architecture that creates a plan and then concurrently executes the plan with multiple search agents. Workflow: Task → Research Director → multiple search agents → final summary by summarization agent. Stay tuned for the release ↓

swarms

17,729 次观看 • 1 年前

Build and test applications faster with a multi-agent coding system. Gain direct experience with Google Antigravity at a Build with AI event and move from design to execution using an IDE with built-in gen AI. 🔹 Multi-agent coding system and IDE 🔹 Lab testing and app development 🔹 High model quotas Get started →

Build and test applications faster with a multi-agent coding system. Gain direct experience with Google Antigravity at a Build with AI event and move from design to execution using an IDE with built-in gen AI. 🔹 Multi-agent coding system and IDE 🔹 Lab testing and app development 🔹 High model quotas Get started →

Google for Developers

15,253 次观看 • 2 个月前

🚀 New LangChain Academy Course: Building Reliable Agents 🚀 Shipping agents to production is hard. Traditional software is deterministic – when something breaks, you check the logs and fix the code. But agents rely on non-deterministic models. Add multi-step reasoning, tool use, and real user traffic, and building reliable agents becomes far more complex than traditional system design. The goal of this course is to teach you how to take an agent from first run to production-ready system through iterative cycles of improvement. You’ll learn how to do this with LangSmith, our agent engineering platform for observing, evaluating, and deploying agents. Enroll for free ➡️

🚀 New LangChain Academy Course: Building Reliable Agents 🚀 Shipping agents to production is hard. Traditional software is deterministic – when something breaks, you check the logs and fix the code. But agents rely on non-deterministic models. Add multi-step reasoning, tool use, and real user traffic, and building reliable agents becomes far more complex than traditional system design. The goal of this course is to teach you how to take an agent from first run to production-ready system through iterative cycles of improvement. You’ll learn how to do this with LangSmith, our agent engineering platform for observing, evaluating, and deploying agents. Enroll for free ➡️

LangChain

30,744 次观看 • 4 个月前

ADK FTW! Learn how to go from >prompt_to agent in this Agent Development Kit (ADK) demo. Our new #OpenSource framework simplifies the process of building agents and sophisticated multi-agent systems while maintaining precise control over agent behavior ↓

ADK FTW! Learn how to go from >prompt_to agent in this Agent Development Kit (ADK) demo. Our new #OpenSource framework simplifies the process of building agents and sophisticated multi-agent systems while maintaining precise control over agent behavior ↓

Google Cloud Tech

10,428 次观看 • 1 年前

Claude Code sub-agents is probably the easiest way to build complex, custom agentic systems. Not just for code, but for anything you can imagine. Watch me build a multi-agent deep research system with Claude Code subagents. Chain subagents with /commands for reliability.

Claude Code sub-agents is probably the easiest way to build complex, custom agentic systems. Not just for code, but for anything you can imagine. Watch me build a multi-agent deep research system with Claude Code subagents. Chain subagents with /commands for reliability.

elvis

190,039 次观看 • 11 个月前

The future of embodied AI revolves around *collaborative* multi agent scenarios that need natural language communication, task delegation, resource sharing, and more ⛏️ Here are MINDcraft and MineCollab, a simulator and benchmark purpose built to enable research in this area!

The future of embodied AI revolves around collaborative multi agent scenarios that need natural language communication, task delegation, resource sharing, and more ⛏️ Here are MINDcraft and MineCollab, a simulator and benchmark purpose built to enable research in this area!

Prithviraj (Raj) Ammanabrolu

33,931 次观看 • 1 年前