I’m a Software Engineer specializing in Voice AI and conversational AI, with hands-on experience building production Voice AI Agents using STT, TTS, LLM orchestration, and real-time streaming pipelines. I focus on solving core voice challenges such as latency, turn-taking, interruption handling, context retention, and reliable conversation flow to build natural and scalable AI experiences. At Abby Connect, I build and enhance multilingual English/Spanish Voice AI agents that automate front-office calls, answer FAQs, handle complex user interactions, and book appointments through backend scheduling systems. My work spans conversation state management, turn-taking, interruption recovery, overlapping speech, LLM function calling, prompt engineering, deterministic guardrails, and human-in-the-loop fallbacks to ensure reliable production conversations. I also work on improving voice-agent quality through automated AI-vs-AI evaluations and scenario-based test suites covering STT, TTS, language routing, and conversation orchestration. I’ve implemented background-noise handling, speaker diarization, and pronunciation normalization. Beyond the voice layer, I build full-stack AI features using React, FastAPI, PostgreSQL, and Salesforce integrations, working closely with Product, Customer Success, and GTM teams to translate real customer workflows into production features. I’m particularly interested in Voice AI, conversational AI, agentic systems, real-time speech technology, LLM applications, AI evaluation, and full-stack engineering for AI-powered products. I’m currently open to full-time opportunities in Voice AI, Conversational AI, and Agentic Systems. Feel free to connect or reach out at
siddanthapusravani@gmail.com
