Get live statistics and analysis of Virat Singh's profile on X / Twitter

I’ve been building a real-world AI hedge fund. It's open source so you can learn + build too. The hedge fund has 6 agents: 1 • market data agent 2 • quant agent 3 • fundamentals agent 4 • sentiment agent 5 • risk manager agent 6 • portoflio manager agent The hedge fund is powered by @LangChainAI You can run all of my code below. No prior coding experience is required. All of the agents show their reasoning so you can see how they work. 1 • Market Data Agent: gathers market data like stock prices, fundamentals, etc. 2 • Quant Agent: calculates signals like MACD, RSI, Bollinger Bands, etc. 3 • Fundamentals Agent: analyzes profitability, growth, financial health, and valuation. 4 • Sentiment Agent: looks at insider trades to determine insider sentiment. 5 • Risk Manager: determines risk metrics like volatility, drawdown, and more. 6 • Portfolio Manager: makes final trading decisions and generates orders. This is just the start. There are a ton of cool features that we can add. Let me know if you have any suggestions.

Introducing Dexter. An open source financial agent in ~200 lines of code. Think: Claude Code, but for finance. What Dexter does today: • plans tasks • runs on its own • validates its work • researches stocks It’s built with OSS tools like @LangChainAI for managing LLMs, prompts, outputs, etc. No coding experience needed to run. Goal: build the best financial agent and share every line of code along the way. Next: evals, starting with @_valsai

Introducing Dexter 2.0 Open source. Built for financial research. Like Claude Code, but for stocks. What Dexter does: • plans tasks • runs on its own • validates its work • researches stocks It uses OSS tools like @LangChainAI, with a fresh stack of typescript, react,…

I made a diagram of my AI hedge fund. It's a multi-agent, multi-LLM system. Thanks to @LangChainAI, it’s evolved a ton. Our AI agents: 1 • Bill Ackman agent 2 • Warren Buffett agent 4 • Fundamentals agent 5 • Sentiment agent 6 • Technicals agent 7 • Valuation agent You can pick any combo of agents to run. Our LLM providers: 1 • OpenAI 2 • Anthropic 3 • Deepseek 4 • Meta We have multiple LLMs from each provider. No coding is required to run the system. And everything is open source. Here is how it works: 1 • You specify tickers to analyze 2 • You choose which agents to run 3 • You choose which LLM to use And that's it. The system takes care of the rest and generates trading signals. If you see an agent that you think is missing, we can add it. If you see an LLM provider that is missing, we can add it. This is just the start, tons more to build!

I read up on DeepSeek’s learning algo, GRPO. GRPO: group relative policy optimization How GRPO works: 1 • model generates a group of answers 2 • compute score for each answer 3 • compute avg score for entire group 4 • compare each answer score to avg score 5 • reinforce model to favor higher scores Other methods like PPO, use a value function model to do reinforcement learning. GRPO does not, which reduces memory and computational overhead when training. A concrete example of GRPO in action: Query: “What is 2 + 3?” Step 1: LLM generates three answers. 1. “5” 2. “6” 3. “2 + 3 = 5” Step 2: Each answer is scored. 1. “5” → 1 points (correct, no reasoning) 2. “6” → 0 points (incorrect) 3. “2 + 3 = 5” → 2 points (correct, w/ reasoning) Step 3: Compute avg score for entire group. Avg score = (1 + 0 + 2) / 3 = 1 Step 4: Compare each answer score to avg. 1. “5” → 0 (same as avg) 2. “6” → -1 (below avg) 3. “2 + 3 = 5” → 1 (above avg) Step 5: Reinforce LLM to favor higher scores. 1. Favor responses like #3 (positive) 2. Maintain responses like #1 (neutral) 3. Avoid responses like #2 (negative) This process is repeated, allowing the model to learn and improve over time. Great work by the @deepseek_ai team.

I built a stock market API in 42 days. All during @_buildspace s5. What is the API? Today, it lets you pull financials for 16,000 tickers going back 30+ years. Why did I build the API? Because the big providers have: • poor API design • poor documentation • expensive sales contracts • strict data retention policies How did I build it? I started building the API from scratch at the start of s5 in early June. I got deep into the SEC Edgar API and cracked how to scrape at scale. Sonnet 3.5 from @AnthropicAI was a godsend for this. I hacked early mornings, late nights, and weekends, with a full-time job. Day by day. Code by code. I set weekly goals and hit each one. End result: https://t.co/Li4F50LrMv I’m excited to see where this API goes. It may go nowhere, but I’ll always know that I can set goals and crush them.

One week since launching Dexter. An open source financial agent in ~200 lines of code. Basically: Claude Code, but for finance. What Dexter does: • plans tasks • runs on its own • validates its work • researches stocks I’m sharing all of our code, so we’re building on OSS tools like @LangChainAI only. As of today, Dexter can read SEC filings and access real-time market data. Next week: deeper agents, evals, and more.

I’ve been building a real-world AI hedge fund. It's open source so you can learn + build too. The hedge fund has 6 agents: 1 • market data agent 2 • quant agent 3 • fundamentals agent 4 • sentiment agent 5 • risk manager agent 6 • portoflio manager agent The hedge fund is powered by @LangChainAI You can run all of my code below. No prior coding experience is required. All of the agents show their reasoning so you can see how they work. 1 • Market Data Agent: gathers market data like stock prices, fundamentals, etc. 2 • Quant Agent: calculates signals like MACD, RSI, Bollinger Bands, etc. 3 • Fundamentals Agent: analyzes profitability, growth, financial health, and valuation. 4 • Sentiment Agent: looks at insider trades to determine insider sentiment. 5 • Risk Manager: determines risk metrics like volatility, drawdown, and more. 6 • Portfolio Manager: makes final trading decisions and generates orders. This is just the start. There are a ton of cool features that we can add. Let me know if you have any suggestions.

I quit my job today. Everything I’ve shared on X over the last 3 years was built in my spare time. Early mornings, late nights, and weekends. Now I’m all in. My focus: 1 • scale @findatasets to $100M ARR 2 • make dexter the best OSS financial agent Grateful for the chance to do this. Let’s build.

I made a diagram of my AI hedge fund. It's a multi-agent, multi-LLM system. Thanks to @LangChainAI, it’s evolved a ton. Our AI agents: 1 • Bill Ackman agent 2 • Warren Buffett agent 4 • Fundamentals agent 5 • Sentiment agent 6 • Technicals agent 7 • Valuation agent You can pick any combo of agents to run. Our LLM providers: 1 • OpenAI 2 • Anthropic 3 • Deepseek 4 • Meta We have multiple LLMs from each provider. No coding is required to run the system. And everything is open source. Here is how it works: 1 • You specify tickers to analyze 2 • You choose which agents to run 3 • You choose which LLM to use And that's it. The system takes care of the rest and generates trading signals. If you see an agent that you think is missing, we can add it. If you see an LLM provider that is missing, we can add it. This is just the start, tons more to build!

Introducing Dexter 2.0 Open source. Built for financial research. Like Claude Code, but for stocks. What Dexter does: • plans tasks • runs on its own • validates its work • researches stocks It uses OSS tools like @LangChainAI, with a fresh stack of typescript, react,…

I built a stock market API in 42 days. All during @_buildspace s5. What is the API? Today, it lets you pull financials for 16,000 tickers going back 30+ years. Why did I build the API? Because the big providers have: • poor API design • poor documentation • expensive sales contracts • strict data retention policies How did I build it? I started building the API from scratch at the start of s5 in early June. I got deep into the SEC Edgar API and cracked how to scrape at scale. Sonnet 3.5 from @AnthropicAI was a godsend for this. I hacked early mornings, late nights, and weekends, with a full-time job. Day by day. Code by code. I set weekly goals and hit each one. End result: https://t.co/Li4F50LrMv I’m excited to see where this API goes. It may go nowhere, but I’ll always know that I can set goals and crush them.

Introducing Dexter. An open source financial agent in ~200 lines of code. Think: Claude Code, but for finance. What Dexter does today: • plans tasks • runs on its own • validates its work • researches stocks It’s built with OSS tools like @LangChainAI for managing LLMs, prompts, outputs, etc. No coding experience needed to run. Goal: build the best financial agent and share every line of code along the way. Next: evals, starting with @_valsai

I made a diagram of my AI hedge fund team We have 4 agents: • portfolio manager • fundamental analyst • technical analyst • sentiment analyst We use LangGraph branching for easy parallel runs. This system can access data for 30,000+ tickers using @findatasets Portfolio manager delegates work to analysts and summarizes their research. Fundamental analyst gets financial statements and analyzes fundamentals. Technical analyst gets stock prices. I will add price pattern tools here. Sentiment analyst gets insider trades and surfs the web for news. This is a simple system, but a good foundation for customizations. Hit me with questions and requests.

I’ve been building in public for 1 year now. Last year, I promised to build and tweet every day for a year. Today, I fulfilled that promise. Year 1 review: • grew twitter from 110 to 11.7K • built 400 member discord community • got 350 stars on a github project • earned first $ from a side project I am naturally an introvert. The toughest part has been tweeting every day. I may miss a day, but as @sahilbloom says, I never miss two in a row. Excited to build and share more in year 2.

It's live. I launched my stock market API for AI financial agents today. • 16,857 tickers • 30+ years of fundamental data Everything is usage-based, meaning no contracts or subscriptions. We have 3 endpoints today: • GET income statements • GET balance sheets • GET cash flow statements All data is directly from the edgar API. You may save any data that you receive from my API. No restrictions. Three things: • create an account 🙏 • send API requests 🚀 • come hang in discord ❤️ I'm just getting started. Excited to help build the future of AI + financial data. cc @_buildspace @_nightsweekends @FarzaTV

I updated my AI hedge fund system diagram. We now have 5 investor agents and 8 LLMs. @LangChainAI powers our agent graph. Our AI agents: 1 • Ben Graham agent 2 • Bill Ackman agent 3 • Cathie Wood agent 4 • Charlie Munger agent 5 • Warren Buffett agent You can mix + match these with any LLM. Our LLM providers: 1 • Anthropic 2 • DeepSeek 3 • Meta 4 • OpenAI 5 • Google (coming soon) 6 • Mistral (coming soon) No coding is required to run the system. Everything is open source. Here is how each investor agent works: 1 • Ben Graham: Uses classic principles like earnings stability, financial strength, and margin of safety with Graham Number calculations. 2 • Bill Ackman: Identifies durable competitive advantages, consistent cash flows, and strong financial discipline at a discount to intrinsic value. 3 • Cathie Wood: Identifies disruptive innovators, focusing on R&D intensity, expanding margins, and breakthrough technologies. 4 • Charlie Munger: Prioritizes business moats, management quality, and predictability while demanding a significant margin of safety. 5 • Warren Buffett: Focuses on owner earnings, consistent growth, and intrinsic value calculations with a long-term perspective on quality businesses. Goal is to keep adding agents and LLMs. Let me know what else to add.