Vetted Retrieval-Augmented Generation Professionals

“Built and productionized an LLM-powered internal knowledge search system in a regulated environment, using embeddings/vector DB retrieval with strict grounding and confidence gating to reduce hallucinations. Reported ~45% accuracy improvement over keyword search and implemented end-to-end orchestration, monitoring, CI/CD, and incremental re-indexing to manage latency and data freshness while driving adoption with business stakeholders.”

“Built and shipped a production RAG system (“TROJAN KNOWLEDGE”) for answering questions over technical PDFs, using a 3-stage retrieval stack (BM25 + FAISS + cross-encoder) to lift F1 from 71% to 84%. Drove major performance gains with a 3-level cache (memory/Redis/disk) cutting latency from ~200ms to ~10ms, and added Prometheus/Grafana monitoring plus LangChain-based fallback logic to handle OpenAI rate limits under load.”

“Software engineer with JPMorgan Chase experience building a real-time operations console backend on Spring Boot/Kafka/Kubernetes and resolving peak-load latency through profiling, indexing, caching, and async processing. Also built and owned an AI-driven digital-archives metadata pipeline during a master’s at UNT using OCR + LLaMA-based prompting with validation, near-human accuracy, and human-in-the-loop guardrails.”

“AI/backend engineer who fine-tuned and deployed a production LLM chatbot using a LangChain + FAISS RAG pipeline, improving latency with PEFT/LoRA and driving strong business impact (40% customer adoption; 92% satisfaction). Also served as technical lead on a data aggregation system for underwriting/quoting, introducing GraphQL for more efficient, maintainable querying and applying CDC to keep cached ranking data fresh at scale.”

“Built a production RAG chatbot for Worcester Polytechnic Institute that indexes 500+ webpages using FAISS + Llama 3, with strong grounding/hallucination controls (confidence thresholds and citations). Also has internship experience orchestrating multi-step ETL pipelines with AWS Step Functions and delivered a 30x faster fraud/claims triage workflow at Munich Re using association rules and stakeholder-friendly dashboards.”

“Built and deployed a production RAG pipeline at PNC Financial Services to let risk/compliance analysts query millions of internal financial documents in natural language, reducing manual search and speeding regulatory validation. Demonstrates deep practical experience with large-scale document ingestion/OCR cleanup, retrieval performance tuning (hierarchical indexing, caching), and LLM reliability controls (grounding, citations, abstention), plus cloud orchestration on Azure and AWS.”

“AI/LLM engineer with healthcare domain experience who built a production clinical support “chart bot” for Molina, including PHI-safe ingestion of 200k+ PDF policies, vector retrieval, and a fine-tuned LLaMA served via vLLM on ECS Fargate. Demonstrated measurable performance wins (HNSW + namespace partitioning; 30% inference latency reduction) and a rigorous evaluation/monitoring approach, while partnering closely with nurses and operations teams to shape workflows and guardrails.”

“LLM/agent engineer at USAA who built a production GPT-4o RAG conversational assistant for financial analysts, focused on regulatory interpretation and internal documentation search. Emphasizes compliance-grade reliability with strict grounding, safe fallbacks, and full auditability via MLflow/DVC plus human-in-the-loop review; reports ~45% reduction in ticket resolution time.”

“LLM/agent production specialist with strong customer-facing and pre-sales chops: turns demo-grade prototypes into reliable, compliant deployments using RAG tuning, guardrails, evals in CI, and observability with staged rollouts/rollback. Known for engineering-first workshops (including live break-and-fix on retrieval misses, tool timeouts, and prompt injection) that win over skeptical senior developers and drive adoption.”

“Backend engineer with insurance/claims domain experience who modernized legacy claims processing systems to support AI-assisted claim review. Emphasizes production-ready API design in Python/FastAPI (schemas, async, caching, graceful degradation), strong observability with Prometheus, and layered security including JWT auth plus database row-level security (Supabase/Postgres).”

“Built a production AI-powered customer support Q&A system using an internal knowledge base to reduce repetitive ticket work and improve customer satisfaction, with an emphasis on source-backed answers and expert oversight. Also has experience defining deployment services in a microservices architecture and integrating large-scale APIs (including work connected to US HHS/COVID-19).”

“Backend-focused Python engineer who has owned production FastAPI services deployed on Kubernetes, including CI/CD (GitLab CI to ECR) and GitOps delivery via ArgoCD/Helm. Has hands-on experience with complex reliability and infrastructure work—solving data inconsistency with validation/partial-data paths, fixing K8s liveness issues via lazy loading, and supporting a phased cloud-to-on-prem migration with dual-writes and monitoring. Also built Kafka-based real-time ingestion consumers handling bursty, high-throughput traffic with async processing and topic/retention tuning.”

“Built and deployed an LLM-powered mental health therapy assistant at AppHealth that segments users by stress level and delivers personalized, non-medical guidance. Implemented healthcare-focused safety guardrails (secondary LLM output filtering) and a multi-agent router workflow validated via statistical tests and therapist review, then scaled training/inference on AWS (EC2/Lambda/DynamoDB) with Kubernetes.”

“Built and deployed a production LLM-powered support assistant for customer support agents using a RAG architecture over internal docs and past tickets, with human-in-the-loop review. Demonstrates strong applied LLM engineering focused on real-world constraints (hallucinations, latency, cost) using routing to smaller models, reranking, caching, and rigorous evaluation/monitoring (offline eval sets, A/B tests, KPI tracking).”

“At FiVerity, built and deployed a production LLM/RAG-based Information Gathering Tool for credit union fraud analysts that generates auditable investigation summaries from verified evidence. Focused on high-stakes constraints—hallucination prevention, cross-entity leakage controls, compliance/PII-safe monitoring, and latency—while also shipping customer-facing agentic workflows using CrewAI and LangGraph in close partnership with fraud and compliance stakeholders.”

“AI/LLM engineer at GPT integrators who built a production multi-agent enterprise workflow integration system, tackling hard problems in agent orchestration, layered memory, and custom RAG over enterprise/user data. Also built an education-focused agent solution integrating with Canvas, Zoom, and email to automate classroom admin tasks, and is currently applying agentic AI to insurance underwriting workflows in collaboration with underwriters.”