Vetted Retrieval-Augmented Generation Professionals

“Software engineer who has owned and improved a customer-facing quote flow in a Vue/Nuxt app, using production observability to reduce latency and improve reliability via caching and request-handling fixes. Also shipped an internal LLM Q&A tool using embeddings + RAG over approved company docs and past support tickets, with guardrails, logging, and an evaluation loop that drove retrieval/prompt improvements. Seeking ~$110k base and requires H1B transfer sponsorship.”

“Built a production LLM/RAG document analysis system for large financial documents (credit reports/PDFs) to help business analysts extract insights faster. Implemented end-to-end pipeline orchestration with LangChain, vector search (e.g., FAISS), and hallucination controls (context grounding, similarity thresholds, and no-answer fallback), delivered as a Dockerized Python API.”

“Built and shipped a production RAG assistant using GPT-4, LangChain, and Pinecone/FAISS to search 50K+ institutional documents, with a strong focus on groundedness and hallucination reduction through retrieval optimization and re-ranking. Pairs this with a metrics-driven evaluation/monitoring approach (BLEU/ROUGE, manual sampling, logging) and workflow automation via Airflow, and has experience translating stakeholder needs into iterative AI prototypes.”

“Built and monetized an object-identification app end-to-end (FastAPI backend, HTML/JS frontend, SQLite→Postgres, auth, and an iOS wrapper via Capacitor/Xcode with Apple privacy/policy compliance). Also productionized an AI-native geospatial metadata/QA assistant using LLM+RAG plus deterministic Python validation, measuring impact via time-to-first-pass review and rework rate, and has experience modernizing legacy GIS workflows and delivering across USDA/FEMA-style teams with disciplined Jira-based execution.”

“Senior full-stack engineer with strong healthcare domain experience who has shipped an Azure OpenAI RAG-based patient medication support chatbot to production, driving ~10K queries/month and a reported 38% reduction in call center volume. Also builds polished real-time React/TypeScript pharmacy tooling and operates large-scale Python/Spark ETL pipelines (~12M records/day) with strong API design, observability, and cloud deployment experience across Azure/Kubernetes and AWS.”

“Robotics engineer who built a smart campus tour robot on a Kobuki Turtlebot using ROS 1, implementing a full navigation stack (semantic world model, A* planner, tour executor, path follower) and integrating SLAM (gmapping) plus a hybrid reactive safety controller. Experienced taking systems from Gazebo simulation to real hardware, including extensive real-world debugging and Docker-based development to handle ROS/Ubuntu version constraints; planning a move to ROS 2 on Turtlebot 4.”

“Backend engineer with hands-on experience scaling a CVE processing platform by re-architecting it into a Kafka-based distributed system, boosting throughput to 200k+ records/min while designing for HA, deduplication, and fault tolerance. Also led a Flyway-driven migration affecting 15M+ records with staged dev→stage→prod rollout, and has implemented production security patterns (Auth0, OAuth2/HTTPS, AWS IAM RBAC) including least-privilege hardening.”

“Backend engineer who built an AI-driven "Smart Feedback Analyzer" API (Flask → FastAPI) that processes user feedback with NLP (Hugging Face + OpenAI) and returns structured insights. Demonstrates strong production-minded architecture: stateless services, Cloud Run + Docker deployment, Redis/Celery background processing, and Postgres/SQLAlchemy performance tuning (EXPLAIN ANALYZE, indexing, N+1 fixes), plus multi-tenant data isolation via JWT/API-key derived tenant IDs.”

“LLM/agentic systems builder who shipped production workflows at Angel Flight West and Eureka AI, combining LangGraph + RAG (Postgres/pgvector) with strong observability (LangSmith/Langfuse). Delivered large operational gains (address lookup cut from 10 minutes to 60 seconds; accuracy to 92%) and has a track record of quickly stabilizing customer-critical pipelines (Pydantic-enforced JSON for ETL) while partnering with sales/ops to drive adoption.”

“AI Automation Engineer at BalancedTrust who single-handedly shipped production LLM features for FinTech compliance: a policy gap-analysis pipeline (SOC 2/GDPR) and a RAG-based regulatory chatbot. Deeply focused on reliability in high-stakes legal/compliance settings, with strong production engineering (edge functions, parallelized batching to cut latency, structured JSON outputs, guardrails, and monitoring) and close collaboration with non-technical compliance experts.”

“Built and deployed a domain-specific LLM chatbot for research/support, cutting manual effort by ~50%. Demonstrates strong applied LLM engineering: RAG, prompt grounding with citations and fallbacks, embedding/top-k tuning, and production monitoring (confidence, latency, feedback loops). Experienced orchestrating agent workflows with LangChain-style pipelines and continuous evaluation to maintain reliability.”

“AI/ML engineer with experience at Northern Trust and Persistent Systems building production LLM + RAG systems for regulated financial use cases, including liquidity forecasting, anomaly detection, and credit scoring. Emphasizes compliance-first design with explainability (SHAP), traceability (MLflow), and hallucination controls (FAISS + citation-grounded prompting), and has delivered drift-triggered retraining pipelines using Airflow and Kubernetes while translating model outputs into business-ready marketing segments.”

“Built and shipped a production LLM-powered financial Q&A interface that extracts precise numeric data from PDFs using a hybrid AWS Textract + LLM normalization pipeline, with confidence gating and guardrails to prevent unreliable answers. Experienced with LangChain-based RAG orchestration (chunking, memory, structured outputs) and collaborated closely with PMs/analysts on IRS Form 990 extraction requirements.”

“Built and deployed a production LLM/RAG clinical document understanding and summarization system for healthcare, focused on reducing manual review time while meeting strict accuracy, latency, and compliance needs. Demonstrates strong MLOps/orchestration depth (Airflow, Kubernetes, Azure ML Pipelines) and a rigorous approach to hallucination mitigation through layered, source-grounded safeguards and stakeholder-driven requirements with physicians/compliance teams.”

“Built and deployed JobMatcher AI, an LLM-driven workflow automation product for job seekers that extracts requirements from job descriptions, matches to user skills, and generates tailored outreach. Demonstrated strong production engineering by cutting per-run cost ~70%, improving reliability with retries/backoff/fallbacks, and reducing hallucinations via schema validation and templating; also orchestrated the system with LangGraph plus Docker Compose across API, vector DB, and workers.”

“Graduate researcher who builds production-grade AI systems spanning LLM security evaluation and on-device RAG. Created HoneyLearner, a self-learning attack framework using GPT-4-class models as structured black-box attackers against honeywords defenses, with rigorous metrics and reproducible orchestration (Airflow/Spark/Kafka/Docker). Also partnered with agriculture scientists at Texas A&M–Corpus Christi to deliver UAV + 3D point-cloud crop-stress maps that cut time-to-insight ~40% and enabled ~30% earlier interventions.”

“Backend engineer who built an AI RAG quoting system for the fastener industry, reducing quote turnaround from weeks to ~30 minutes and raising retrieval accuracy to ~90% by solving a semantic-collision issue with a parent-document retrieval design. Strong in production AWS integrations (Cognito auth, S3 pre-signed uploads), performance optimization (multithreading/out-of-core), and real-time streaming (Kafka/Spark Kappa architecture achieving sub-second latency), plus Kubernetes logging and GitHub Actions CI/CD to ECR.”

“Built and deployed an enterprise RAG application at Centene to help clinical teams retrieve insights from large internal policy document sets, cutting manual research by 30–40%. Implemented custom domain-adapted embeddings (SageMaker + BERT transfer learning) and hybrid retrieval (BM25 + Pinecone) to drive a 22% relevance lift, and ran the system in production on AWS EKS with CI/CD, MLflow, and Prometheus monitoring (99% uptime, ~40% latency reduction).”

“AI/ML engineer (Merkle) with hands-on experience deploying RAG-based LLM applications and real-time recommendation engines into production. Strong in cloud/on-prem architectures, GPU autoscaling, caching, and network optimization—delivered measurable latency reductions (40–70%) and improved retrieval relevance by systematically benchmarking chunking/embedding configurations and validating pipelines via CI/CD.”

“Healthcare data/NLP practitioner with experience at UnitedHealthcare building production ML systems that connect unstructured call center transcripts and medical notes to structured claims data. Has delivered measurable impact (25% classification accuracy lift; ~30% relevance improvement) using classical NLP, embeddings (Sentence-BERT + FAISS), and AWS SageMaker deployments with robust validation and drift monitoring.”

“LLM solutions practitioner with SOC/alert-triage experience who takes LLM prototypes to production using RAG (Pinecone), FastAPI services, guardrails, CI/CD, monitoring, and robust fallback logic. Known for rapid real-time debugging of embedding/vector and agent workflow issues, and for driving adoption through code-first workshops and sales-aligned custom demos with measurable improvements (35% faster triage; 40% increase in correct tool usage).”

“Built and shipped Serrano AI, a multi-tenant SaaS conversational AI platform that automates Odoo ERP workflows and lets ops/finance/supply-chain teams query ERP data in natural language. Implemented a multi-agent architecture (LangChain/LangGraph/CrewAI) with hybrid RAG over ERP schemas, deployed on Heroku/Vercel with production observability, cutting reporting time by ~80% while addressing hallucinations, latency, and schema complexity.”

“Built Hexagon’s production Text-to-CAD Copilot that converts text and rough sketches into editable CAD code, combining GraphRAG (Neo4j/LangChain) with a Gemini-powered vision module and multi-agent geometric validation—cutting manual modeling from a day to ~45 seconds and driving retrieval latency below 50ms. Also has large-scale GCP data/ML orchestration experience (Airflow/Cloud Composer, Dataflow, Pub/Sub, Snowflake) processing 50M+ daily records with drift monitoring and automated reliability controls.”

“Full-stack engineer with Deloitte and CrowdDoing experience shipping production web platforms on AWS (EC2/RDS/S3/Fargate) using React/TypeScript and Node/Express/PostgreSQL. Built customer-facing authentication/SSO flows (OAuth2 + JWT) and state-specific US privacy consent workflows, and also delivered a Python/Flask LLM-based finance document parser chatbot with vector DB integration and latency optimizations.”