Vetted Model Evaluation Professionals

“AI/back-end engineer at Canon who helped build and operate an internal production LLM platform that acts as a secure middle layer between users and models, defending against jailbreaks/prompt injection while enabling RAG, memory, and grounded responses over company data. Experienced with LangChain/LangGraph orchestration, vector DB retrieval, and reliability practices (testing, monitoring, adversarial prompts) to run high-throughput, low-latency AI workflows in production.”

“Built and deployed LLM-powered features for a startup organizational management application, focusing on real-world deployment constraints like latency and cost. Implemented RAG with FAISS and improved retrieval quality by switching embedding models (OpenAI/Hugging Face) and fine-tuning embeddings on medical corpora for a medical-report UI feature. Uses LangChain and LangGraph to orchestrate multi-node LLM API workflows and evaluates systems with metrics like latency, cost per request, and error taxonomy.”

“Built and shipped a production Python-based agentic RAG document retrieval system over 80K records using FastAPI, OCR, vector search, and AWS infrastructure, with a strong emphasis on reliability, testing, and observability. Stands out for treating AI failures like production incidents—turning hallucinations, retrieval misses, and OCR issues into regression tests—and for quantifiably reducing document lookup time from about 12 minutes to under 90 seconds.”

“Built and deployed a production LLM-powered text-to-SQL/document intelligence chatbot on AWS that lets non-technical business users query complex enterprise databases in plain English. Demonstrates deep practical expertise in schema-aware prompting, embeddings-based schema retrieval, SQL safety/validation guardrails, and rigorous offline/online evaluation with human-in-the-loop approvals for risky queries.”

“Analytics-focused candidate with hands-on project experience in SQL data preparation and Python-based churn modeling. They demonstrated a practical approach to turning messy multi-source data into reporting tables, validating data quality rigorously, and translating churn insights into targeted retention strategies.”

“Built a production-grade, semi-automated document recognition and classification system for large volumes of scanned PDFs, starting from little/no labeled data and handling highly variable scan quality. Deployed on AWS using SageMaker + Docker and orchestrated on EKS with a microservices design that scales CPU-heavy OCR separately from GPU inference, with strong reliability controls (validation, fallbacks, retries, readiness probes).”

“Data/ML practitioner who has built a production RAG-based knowledge assistant integrated into Microsoft 365/internal dashboards to help employees query internal documents in plain English. Experienced orchestrating and hardening ETL pipelines with Airflow and Azure Data Factory (validation, retries, monitoring) and running end-to-end model evaluation and production performance tracking via Power BI.”