hi theređź‘‹, I'm

Edo Fransisco

20, he/him

BLOCKCHAIN DEVELOPER | LLM & PROMPT ENGINEER

about me.

Results-driven LLM Engineer and AI Agent Developer specializing in multi-agent orchestration, tool-augmented LLMs, and automated research/customer service solutions. Delivered real-world impact for global clients—streamlining workflows, reducing costs, and shipping production-ready AI systems. Seeking remote roles where I can drive the next generation of AI automation.

experience.

  • Full-Stack & Blockchain Developer Remote

    at, independent-contractor

    2023 - 2025

    • Shipped 10+ DApps to Ethereum/Polygon (Hardhat, Ethers.js, IPFS); 99.9% uptime and 10K+ monthly users.
    • Decreased audit-found vulnerabilities by 40% using OpenZeppelin, Hardhat, and custom test frameworks.
    • Built and maintained RESTful APIs, secure blockchain voting DApp (1,000+ DAU), and real-time dashboards (React.js, Next.js).
    • Delivered web3 and DeFi solutions for clients in fintech and voting sectors
    • MERN Stack
    • Next.js
    • TypeScript
    • Solidity
    • Javascript
    • Python
    • Ethereum
    • Web3.js
    • Ethers.js
    • Metamask
    • Hardhat
    • Truffle
    • OpenZeppelin
    • AWS
    • Google Cloud Platform (GCP)
    • Vercel
  • LLM & AI Agent Developer Remote

    at, independent-contractor

    2024 - Present

    • Delivered production-grade LLM automations and multi-agent systems for e-commerce and fintech clients under NDA, driving up to 40% faster research workflows and 30% improvement in customer response accuracy.
    • Built and A/B tested 100+ advanced prompts (ChatGPT, Claude, Grok) using synthetic datasets and OpenAI evals; optimized resolution rates for high-volume chatbot deployments.
    • Engineered robust role-based dialog flows and RAG-powered customer support agents with OpenAI GPT-4, LangGraph, CrewAI, and Qwen; reduced query handling time by 25%.
    • Developed modular, tool-using conversational agents (LangChain, Pinecone, custom API integrations); improved session continuity and minimized repeated queries by 35%.
    • Led full project lifecycle: requirements, design, deployment, and production monitoring-delivering solutions used by 10K+ end users.
    • AI Model Understanding
    • Prompt Optimization
    • Natural Language Processing (NLP)
    • Automation & Workflow Design
    • Client Consultation & Adaptability
    • LangGraph
    • LangChain
    • OpenAI Agent SDK
    • CrewAI
    • OpenAI API
    • Google Gemini
    • Anthropic Claude

projects.

  • CRYPTO TRADER AI AGENTS

    github

    • Automated end-to-end crypto trading plan generation, integrating real-time market data (Binance API) and news sentiment.
    • Sequential multi-agent validation delivers actionable strategies, risk management, and Markdown reports.
    • CrewAI
    • Python
    • Web Scraping
    • Binance API
  • AIRLINE CUSTOMER SUPPORT AGENT

    github

    • Modular agent (Qwen-Agent) integrates real-time flight search/status tools.
    • Features multi-turn dialog, robust prompt flows, and a mock backend for safe, complete testing.
    • Qwen-Agent
    • Python
    • Custom Tools
  • DIGITAL TWIN AI PERSONAL ASSISTANT

    github

    • Engineered a proactive AI that learns a user's persona via RAG, securely integrating with Google Calendar & Gmail to draft context-aware emails in the user's personal style and autonomously suggest actions.
    • Qwen-Agent
    • Python
    • RAG
    • ChromaDB
    • Google API (OAuth 2.0)
  • TokenPicker Agent

    github

    • Automated the initial research phase, reducing manual data collection time by an estimated 70-80% compared to traditional methods.
    • Enhanced research quality by systematically scanning and summarizing data from over 50 diverse sources (e.g., news sites, social media, whitepapers) per token analysis.
    • Streamlined the delivery of insights by generating daily automated summary reports that highlighted key findings and rationales, enabling quicker decision-making for hypothetical investment strategies.
    • Implemented a data validation step within the agent workflow that improved the reliability of curated resources by an estimated 90%, filtering out irrelevant or low-quality information.
    • Python
    • CrewAI
    • OpenAI API
    • Pushover API
    • Serper.dev

technical skills.

AI & Machine Learning

Prompt EngineeringNLPLLMOpenAI APIAI Agent DevelopmentGPTClaudeGrokDeepSeekQwen

Programming

PythonTypescriptSolidityJavaScript

Blockchain

Ethers.jsWeb3.jsHardhatTruffleOpenZeppelin

Web Development

React.jsNext.jsNode.jsExpress.jsHTMLCSS

Cloud & DevOps

AWSGoogle CloudVercelGit

Databases

MongoDBMySQLSQLAlchemy

contact me.

Feel free to reach out to me for collaboration, project inquiries, or just to say hello. I'm always open to discussing new opportunities and ideas.