Praxeon Solutions is where Amna and Zarmina document their hands-on work in applied AI — from retrieval and fine-tuning to the MLOps that gets systems into production.
We're two AI/ML engineers currently interning and exploring full-time roles. Praxeon Solutions is where we document real, hands-on implementation work — retrieval-augmented generation, LLM fine-tuning, and the MLOps that gets systems into production. We're learning by building, and sharing what we find along the way.
Chunking strategy, embedding choices, and retrieval pipelines — with a focus on keeping model outputs grounded and reducing hallucination.
Adapting models for specific tasks, and comparing tradeoffs across cost, latency, and output quality.
Multi-step agent pipelines built with LangGraph, designed with reliability in mind — not just capability.
Model serving, monitoring, and the operational work that keeps AI systems stable once they're live.
AI Engineer Intern focused on retrieval-augmented generation and LLM fine-tuning. Works hands-on with LangChain and LangGraph to build and evaluate retrieval pipelines. Currently exploring full-time roles in applied AI.
LinkedIn ↗ML Intern focused on MLOps and model deployment. Works on containerization, CI/CD, and monitoring to keep ML systems reliable once they're in production. Currently exploring full-time roles in applied AI.
LinkedIn ↗Whether it's a RAG system that needs to stop hallucinating, a model that needs fine-tuning, or an MLOps pipeline that needs to actually ship — we'd like to hear about it.