José Eduardo Bartra Quispe

Python Developer & AI Engineer. With over 10 years of experience in automation with Python and concurrent programming (Java), I specialize in building intelligent systems using LangChain, RAG pipelines, and multi-provider LLM architectures. My work spans from prompt engineering and vector databases to real-time voice-enabled AI agents deployed in production.

Notable Architecture & Projects

MHT-ChatRAG: Real-Time Multilingual Assistant

Python | Node.js | LangChain | RAG | LLMs (GPT, LLaMA, Ollama) | Prompt Engineering | APIs REST | WebSocket | LanceDB

Architected a production-ready RAG system using LangChain with hybrid retrieval (Vector + FTS). Integrated multi-provider LLM fallback chains (Cerebras, OpenAI, Gemini, Ollama) with adaptive prompt engineering. Built REST APIs for configuration management and WebSocket-based real-time streaming with TTS/STT capabilities.

View Case Study →

Adaptive Multi-Source AI Data Manager

  • Designed a system capable of dynamically switching between multiple data providers (e.g., music APIs, game presence, or AI sources).
  • Implemented fallback logic to ensure continuous data availability.
  • Enabled real-time UI updates based on source state (ACTIVE / INACTIVE / OFFLINE).
Key Contributions: Provider abstraction layer, adapter-based data normalization, autonomous source selection logic.

Modular Command & Trigger System

  • Built a reusable command system with centralized registry and distributed execution contexts.
  • Designed multiple managers for handling commands across entities, blocks, and global scopes.
Key Contributions: Unified trigger interface, command reuse via global repository, persistent state handling.

Foundational Experience & Maneuvers

Automation & Concurrent Systems (10+ Years)

Java | Python | SQL | Web Scrapping

  • Thread Management: Deep understanding of concurrent programming in Java, directly applying this logic to modern Node.js Event Loop architectures and LWW concurrency in AI streaming.
  • Autonomous Bots: Developed data-driven Python bots and macros for automated data parsing, web scraping, and specialized assistance.

Conversational UI & Game Dev

Watson Assistant | Tidio | No-Code Platforms

  • Chatbot Design: Orchestrated conversation flows and natural responses using tools like Watson Assistant, laying the groundwork for complex LLM prompting and RAG integrations.
  • Game Dev & Optimization: Directed high-concurrency online game rooms and optimized JSON/multimedia resources, establishing my focus on performance and low-latency architectures.

Core Skills

Python & AI Frameworks

  • Python (Advanced, 2+ years production)
  • LangChain / RAG Pipelines
  • Prompt Engineering & LLM Tuning
  • LLMs: GPT-4, LLaMA, Gemini, Ollama

Backend & APIs

  • APIs REST (Design & Consumption)
  • Node.js / Express / WebSockets
  • Vector DBs: LanceDB, Embeddings
  • Redis / PostgreSQL

DevOps & Tools

  • Docker & Container Orchestration
  • Linux Server Administration
  • Git & CI/CD

Languages

  • Spanish (Native)
  • English (Professional Working)

Value Proposition

  • Building production-ready RAG systems with LangChain and Python.
  • Designing multi-provider LLM architectures with intelligent fallback.
  • Prompt Engineering for complex conversational AI agents.
  • Bridging AI backend with real-time frontend experiences.

Education & Certifications

  • Node.js for Bots (2023 - Present)
    Advanced bot automation and programmatic tooling.
  • Diseño y Desarrollo de Videojuegos (2021 - 2023)
    Instituto San Ignacio de Loyola (ISIL).
  • Videojuegos y Realidad Aumentada (2020)
    SENATI - Technical Training.