Amro Eid

I'm an AI Engineer with around 2 years of professional software and AI engineering experience, focused on building and shipping production AI systems. My main interests are machine learning, LLMs, generative AI, AI agents, RAG, and the engineering needed to make these systems reliable in production. At Neuvox, I work on real-time voice AI systems for customer-facing applications. I've built a production voice pipeline using Python, FastAPI, Pipecat, Twilio WebSockets, and AWS ECS Fargate, connecting speech-to-text, LLM, and text-to-speech components with around 500–900ms end-to-end latency. I've also worked on multi-provider AI infrastructure across OpenAI, Groq, Mistral, Deepgram, ElevenLabs, and Cartesia, as well as AWS Lambda APIs, DynamoDB, S3, SQS, Cognito, Docker, and GitHub Actions. On the ML side, I've worked with fine-tuning and inference for open-source models, including Mistral models with QLoRA and Unsloth, quantized inference with vLLM on A100 GPUs, and RAG pipelines using embeddings and vector databases. Outside work, I build projects on my own, including GeoCodex, an open-source QGIS plugin that lets users interact with PostGIS data using natural language and images. I'm looking for a remote AI/ML or Applied AI role where I can work on real products, take ownership of technical problems, and continue going deeper into LLMs, agents, model evaluation, and production AI systems.