Hire Elijah G. - LLM Engineer - Remote

* Designed and implemented retrieval-augmented generation (RAG) pipelines integrating LangChain, LlamaIndex, and OpenAI API, ensuring factual grounding and reduced hallucination rates in production. * Built and deployed LLM-based microservices using Python, FastAPI, and Docker, orchestrated across Kubernetes clusters for scalable inference. * Created reusable prompt-template systems and evaluation frameworks, improving output consistency and QA pass rate by 30%. * Experimented with agentic architectures (LangGraph, CrewAI, AutoGen) to automate internal documentation and customer workflow handling. * Collaborated with product, creative, and data science teams to integrate LLM workflows into customer dashboards and analytics systems. * Developed observability tooling for LLM telemetry and cost tracking using LangSmith and PromptLayer. * Contributed to a unified MLOps pipeline (ClearML + MLflow) for versioning, dataset governance, and reproducible model releases. * Led internal workshops on prompt engineering, model evaluation, and secure AI deployment across cloud and on-prem environments.

* Engineered end-to-end AI solutions using PyTorch, TensorFlow, and Transformers for multimodal data (text, vision, and tabular). * Built RAG and search systems combining vector databases (FAISS, Pinecone) with LLMs to power internal Q&A and summarization tools. * Deployed Azure-based GenAI microservices for document intelligence, integrating OpenAI API, LangChain, and Azure Cognitive Search. * Developed automated model evaluation pipelines using metrics such as accuracy, relevance, and hallucination rate. * Collaborated with cross-functional teams to implement LLM-powered copilots for content generation and knowledge retrieval. * Built MLOps infrastructure with Docker, GitHub Actions, and Kubernetes, reducing deployment time from hours to minutes. * Implemented prompt optimization and few-shot inference strategies for domain-specific use cases. * Mentored junior engineers on best practices for GenAI deployment, data curation, and secure LLM integration.

* Developed and maintained ML pipelines for classification and NLP applications using scikit-learn, TensorFlow, and FastAPI. * Introduced CI/CD for ML models through Docker, GitLab CI, and AWS SageMaker, improving reproducibility and delivery speed. * Integrated retrieval components into internal chat and search systems, paving the way for early RAG-style applications. * Created automated data preprocessing and labeling workflows using Airflow and PostgreSQL. * Collaborated with DevOps teams to optimize GPU utilization and batch inference throughput. * Designed monitoring dashboards for model drift, performance degradation, and retraining triggers. * Assisted in transitioning legacy ML scripts to containerized microservices, enabling team-wide scalability. * Supported client teams in AI product ideation, transforming business requirements into production ML solutions.