Felipe Pinzon
Solutions Architect - Alexa @Amazon
Signup · Get unlimited contacts
WORK HISTORY
Solutions Architect - Alexa @Amazon
Sunnyvale, CA, US
Built four functional multi-agent AI systems in Python including a real-time voice avatar using LiveKit, ElevenLabs, and Hedra, and RAG agents on AWS Bedrock. VP-reviewed demo presented at a private Alexa+ developer event.* Advised partners on Alexa+ Multi-Agent SDK integration patterns including Model Context Protocol (MCP) and Agent-to-Agent (A2A) protocol, shaping how partners build toward the agentic AI platform.* Guided 20+ worldwide Alexa Built-In partners across TVs, soundbars, hearables, and smart speakers through device integration and certification, spanning multiple product generations and device categories.* Identified the root cause of a critical False Rejection Rate issue in the wake-word engine cascade architecture, drove fixes across Amazon and partner engineering teams, and delivered a measurable improvement in Alexa responsiveness across Built-In devices.* Authored four technical App Notes on AFE tuning, DRM implementation, output speech level calibration, and audio troubleshooting, adopted as implementation references by partners and SoC vendors worldwide. Published on the Alexa Developer Blog.* Originated a cross-functional audio gain staging standardization methodology, authored the 1-pager that fed a director-level certification goal, and delivered partner-facing guidance with tooling shipped in 2024.
EDUCATION
Pontificia Universidad Javeriana
Engineer's degree, Electrical, Electronics and Communications Engineering
University of Notre Dame
Master’s Degree, Engineering and Business
SKILLS
ABOUT FELIPE PINZON
Quality is not a feature. It\'s the difference between a product people tolerate and one they love.I\'ve spent over a decade working with that principle in mind — building things that are actually useful, not just functional. My work sits at the intersection of audio engineering, AI systems, and software development — three domains that rarely overlap, and that I\'ve spent my career combining deliberately.We\'re at an inflection point with voice AI. For the first time, systems can understand natural language the way humans actually speak it — and translate that into real action. I believe voice-first AI assistants have a rare opportunity right now to take away the tasks that consume people\'s time and attention, so they can spend it on what actually matters to them.I build toward that. I\'ve developed multi-agent AI systems with real-time voice capabilities, built Python automation frameworks for audio pipeline validation, and worked hands-on with consumer device partners — always pushing the quality bar beyond what was minimally required. I\'m equally comfortable going deep with a DSP engineer and leading a technical conversation at the C-suite level.The bar for what\'s possible keeps moving. That\'s the point — and exactly what keeps me pushing.Core skills: Python · Audio Signal Processing · DSP · Voice AI · Test Automation · Multi-Agent AI Systems · AWS · Solutions Architecture · AFE/AEC · VoIP
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.