Emanuel – AWS, Python, LLM

Sales teams spent heavy manual effort on cold calling, qualifying leads and logging what happened on each call. Emanuel designed a platform where AI voice agents run outbound campaigns, hold live conversations, and turn every call into structured, CRM-ready results.

• Designed back-end APIs for campaign management, lead management, call execution, and call outcomes.
• Built workflows for AI voice agents to conduct outbound calls using predefined campaign goals and dynamic conversation logic.
• Integrated voice agent infrastructure with telephony providers and AI/LLM services.
• Implemented structured call result extraction, including qualification status, intent, objections, next steps, and follow-up notes.
• Supported call transcript processing, summarization, and CRM-ready outputs.
• Designed validation and guardrails to make AI conversations more reliable and aligned with campaign rules.
• Worked on containerized backend services for production-ready deployment.
• Collaborated on product workflows for campaign setup, monitoring, and performance tracking.

Businesses wanted AI agents that do real work through their existing tools, not just chat. As CTO, Emanuel designed the platform for creating, deploying and orchestrating agents that execute business workflows through APIs, tools and integrations.

• Designed the back-end architecture for agent execution, tools, memory, and integrations.
• Built API services for chat, agent execution, workflow triggers, and external integrations.
• Led cloud deployment and infrastructure decisions.
• Integrated AI agents into automation workflows.
• Designed reusable agent templates and business-specific AI workers.

A Colombian textile company received a mix of retail, wholesale, ecommerce and international enquiries on WhatsApp, often as photos and voice notes. Emanuel built a production agent that understands these inputs, answers product questions and routes each customer to the right sales channel.

• Designed the agent logic for routing customers to physical stores, B2B wholesale, ecommerce, or international channels.
• Built media-handling modules for fabric identification from images, screenshots, and TikTok-style content.
• Added voice note processing and natural voice replies.
• Integrated product catalog lookup, lead capture, contact management, and business rules.
• Improved reliability through validation, fallbacks, and monitoring of production conversations.

It's an AI-driven recruitment platform that replaces traditional CVs with personalized video applications, supporting candidates, recruiters, region leaders, admins, AI feedback, consultation flows, and payment/commission logic.

• Defined the software requirements and system architecture for the product.
• Designed user roles for candidates, recruiters, region leaders, admins, and system-level AI services.
• Created AI workflows for script generation, video feedback, candidate preparation, and recruiter support.
• Designed affiliate/commission structures and payment integration logic.
• Planned scalable data aggregation for hiring insights.