Skip to content
Gustavo PossebonSenior Software Engineer
Zenvia

AI-enhanced chatbot operations and WhatsApp flows

Built and optimized automation products powering chatbot operations, WhatsApp integration, and support efficiency.

Back to case studies
Challenge

What needed to be solved

The product had to scale large interaction volumes while reducing setup complexity, fraud risk, and support friction for operators and customers.

  • High message volume and strict availability expectations.
  • User experience depended on fast response times and dependable flows.
  • AI needed to improve real operational work instead of acting like a novelty layer.
Contributions

What I owned

The emphasis here is on the parts of the work that demonstrate engineering judgment, system thinking, and execution under constraints.

  • Managed automated chatbot flows serving more than one million monthly interactions.
  • Built WhatsApp-based 2FA to reduce fraud and unauthorized access attempts.
  • Led work in the WhatsApp integration squad to improve response times and availability.
  • Introduced OpenAI-powered automation to reduce chatbot setup time and AI-driven chat history retrieval to improve support.
Outcomes

What changed

The strongest evidence in these projects comes from reliability, speed, throughput, and the amount of operational drag removed from teams.

  • Reduced chatbot setup time from two hours to under three minutes.
  • Lowered fraud and unauthorized access attempts by more than 70% within six months.
  • Improved system response time from 4 seconds to 1.2 seconds while sustaining 99.95% availability.
  • Reduced ticket resolution time by 35% with improved support context retrieval.
Takeaway

What this says about how I work

The most valuable AI work is usually invisible in the demo. It shows up as faster setup, cleaner operations, and better decisions for the people using the system every day.