Daniel Barcia Ochoa@danielbo

The Problem In the globalized IT industry, non-native English speakers often struggle during highly technical remote meetings (Zoom, Teams) with international vendors. Native meeting subtitles are often inaccurate for technical jargon and lack real-time translation, while speaking back fluently remains a massive barrier. The Solution: Hermes Hermes is a Real-Time Voice Proxy designed to bridge this gap bidirectionally. Operating completely outside the meeting software via Windows WASAPI loopback and virtual audio cables, Hermes acts as a transparent intermediary without needing bot-invites or API integrations with Zoom. How it works under the hood Listening (The Interpreter): Hermes silently captures the meeting's audio output natively. Using AssemblyAI's cutting-edge Streaming API v3, it transcribes the fast-paced English technical speech with incredible accuracy in real-time. This text is instantly translated to Spanish using Gemini and displayed on a sleek, always-on-top Tkinter overlay. Speaking (The Proxy): When the user holds the "Push-to-Talk" button, Hermes records their native Spanish speech, transcribes it, translates it to precise technical English, and uses advanced Neural TTS to inject a fluent English voice directly into the meeting via a VB-Audio Virtual Cable. Hermes empowers professionals to focus on solving complex IT problems without the cognitive load of language barriers, acting as the ultimate communication copilot.

🚀 The Architecture: "The Brain & The Guardian" Most autonomous trading agents make a critical mistake: they let the LLM directly execute trades and manage capital. During our development, we realized that giving an LLM direct access to liquidity via free-form tool-calling is a recipe for disaster (hallucinations, latency issues, and misunderstood API schemas). Our architecture elegantly separates concerns: The Brain (predictive_brain.py & lobo_scout.py): Scans the market continuously. Feeds live news and technical indicators (RSI, MACD) to the LLM. The LLM's only job is to classify the opportunity and assign a Brain Score (0 to 100). It never touches the money. If the score crosses the threshold (e.g., >65 for Calls, <35 for Puts), a signal is generated. The Guardian (ochoa_core.py): A deterministic, hardcoded Python risk manager. Intercepts the LLM's signals. Checks available portfolio limits, applies Take Profit (e.g., 30%) and Stop Loss limits. Handles the execution securely via Alpaca CLI integration. 🛠️ Key Features & "Build in Public" Milestones Building OCHOA live during the hackathon led to several "Open-Heart Surgeries" on the code, resulting in these advanced features: 1. Alpaca CLI Native Integration 2. The "Zombie Options" Bypass 3. Overdrive (Kamikaze) Mode 4. Dynamic Radar Multi-Bot System

🌎 Hybrid Hackathon | 💻 Online Build: 3–8 November 2026 | On-site Phase: 8 November 2026 📍 On-site Venue: Pierpont Place 163 W Pierpont Ave, Salt Lake City, UT 84101, USA ✨ Please note that travel and accommodation are not covered.

🌎 Online Hackathon | 💻 Month-long challenge | Dates: Sep 1–30, 2026 | 🏆 $10,000 Prize Pool ($5k cash + $5k in AAI credits) A month-long online challenge run by lablab.ai together with AssemblyAI, the Voice AI infrastructure company for builders. Every participant builds on AssemblyAI. You can join at any point — registration stays open for the whole build window, so you can start on day one or pick the project up halfway through and still submit. 🌍 Fully online hackathon. Join and build from anywhere in the world.

🌎 Online Hackathon | 💻 Dates: 28 August–4 September 2026 | 🏆 $6,000 Prize Pool Build AI trading agents on Alpaca — autonomous agents and trading apps using Alpaca's Trading API, MCP server and CLI. 🌍 Fully online hackathon. Join and build from anywhere in the world.