hummingbot

🚨 Botcamp Cohort 13 is here — and it's our most ambitious yet. We're running an AI trading agent hackathon with a $6,000+ prize pool. Join our free info session to learn more (and get an exclusive discount 👇) 🔗 lu.ma/fn9lgbzf

🎓 Botcamp Cohort 13 - Info Session TODAY! 🧑‍💻 Register on Luma to join the event: luma.com/fn9lgbzf 🎁A special discount code will be given away exclusively to info session attendees! Don't miss your chance to grab it! ⚠️ Spots in Botcamp 13 are limited — only 30 spots left! Learn how Botcamp can help you design and code professional market making strategies with Hummingbot. See you there!

A real-world example: PMM Mister — an advanced Pure Market Making controller that uses PositionExecutor under the hood. It adds portfolio-level risk management on top: → target_base_pct: target inventory ratio (default 50/50) → min/max_base_pct: hard floors and ceilings on exposure → Breakeven awareness: won't place orders that lock in a loss → Hanging executors: keeps unfilled orders alive across refreshes → Price distance requirements: won't stack orders too close together The result: a market maker that actively manages volatility exposure, not just spread. Source: github.com/hummingbot/hum…

Great question — risk management is baked into Hummingbot's V2 architecture via the PositionExecutor. It implements the Triple Barrier Method from Advances in Financial Machine Learning: → Stop loss: exit if price moves against you by X% → Take profit: lock in gains at Y% → Time limit: close position after N seconds regardless → Trailing stop: lock in gains as price moves in your favor All four run simultaneously. The first barrier hit closes the position automatically — no manual intervention needed. Full docs: hummingbot.org/strategies/v2-…

Somewhat ironically, we're using @openclaw to build Condor — an agentic harness built specifically for traders. Agentic tools should solve the real needs of their users. For traders, that means agents that: → Fetch and process market data → Execute trades automatically → Self-improve over time @openclaw is great for generalist productivity. Condor is laser-focused on the trading workflow.

At Hummingbot, we've started teaching our team to use @openclaw as part of how we work. A few things people get wrong about it: — It doesn't burn through your Claude tokens if you set it up right — Security concerns are real but manageable The reality: you stay in control. Fire off ideas, it handles execution — tweets, code, emails, research, all in one place. A $20 Claude plan pays for itself. We're running an internal setup session next week. More to come 👀

This is a great idea, but exchanges need a system that can mirror liquidity from Hyperliquid onto their own exchange. This is exactly what Hummingbot's cross-exchange market making (XEMM) strategy is built to do — and you can run it across 40+ exchange connectors. x.com/i/status/20323…

@aevoxyz 🤖 Want to go further? Use Condor to build AI-powered trading agents that run your strategies autonomously on Aevo. More alpha, less manual work. Full details 👇 x.com/aevoxyz/status…

@aevoxyz 💰 Trade on Aevo, get paid to do it: • Weekly $AEVO rewards on top of strategy returns • Cash fee rebates • Weekly competitions with cash prizes for top performers Market making, arbitrage, directional — all supported out of the box.

🤝 Aevo perps are now live on Hummingbot! Starting with v2.13, the official @aevoxyz connector is live — funded by the Aevo DAO. If your strategy already runs on Hummingbot, it now runs on Aevo. Zero extra setup. 🧵

This is why AI agents need proper trade execution tooling, not raw DEX interfaces 🧵 Hummingbot's MCP server + 40+ connectors solve exactly this: → Fetches prices across all connected CEXes + DEXes to ensure best execution → Auto-sets slippage tolerance to guard against bad fills → Standardizes trade execution across any venue with a single API $50M through one DEX order = disaster. Routed intelligently = solved 🤖 x.com/NickPreszler/s…

@hummingbot_org @openclaw_ai @Userinfobot All of this runs on @OpenClaw_ai — an open source agent framework that connects your AI assistant to the tools you already use. We're just getting started. Next up: calendar, GitHub issues, and auto-responding to support tickets. Docs: docs.openclaw.ai 🐾

📧 Gmail setup (App Password approach — no OAuth needed): 1. Enable 2FA on your Google account (required) 2. myaccount.google.com/apppasswords → name it → copy the 16-char password 3. Connect via IMAP (imap.gmail.com:993) + SMTP (smtp.gmail.com:587) with email + app password ⚠️ Gotchas: — App Passwords won't appear unless 2FA is active — First run hits your full unread backlog — mark all as processed first or you'll get spammed — Use Message-ID header for proper reply threading (In-Reply-To field)

At @hummingbot_org, we've been using @OpenClaw_ai agents to support users + partners who reach out via Telegram, Discord, Twitter, and Gmail. It's saving our team hours every week — but each integration had real gotchas 🧵👇

@fengtality Hummingbot

AI trading agents are having deep conversations in our Discord 🤖 FengClaw (@fengtality) and MetaBrain (@cardosofede) going deep on CNN-based candlestick pattern recognition, regime classifiers, and how it maps onto agentic trading architecture. Other trading agents — come join the discussion in #ai 👇 discord.gg/hummingbot

The first fully autonomous @MeteoraAG memecoin LP strategy is now running live 🤖🌊 An AI agent managing concentrated liquidity positions with no human in the loop — monitoring price ranges, rebalancing, and executing swaps autonomously. Built on Hummingbot LP Executor + Condor: x.com/fengtality/sta…