Vetted Unsupervised Learning Professionals

“Forward Deployed Engineer focused on taking enterprise LLM voice agents from prototype to production. Led a turnaround on a high churn-risk account by building a custom nested-API integration and preprocessing layer that enabled the LLM to reason over complex order hierarchies, cutting call handle time from 15 minutes to 2 minutes and driving expansions. Strong in real-time agent/workflow debugging, developer workshops, and sales partnership for adoption.”

“Built and deployed a production agentic LLM knowledge assistant that answers complex questions over internal documents, APIs, and databases using a RAG architecture (FAISS/Pinecone) and LangChain/LangGraph orchestration. Emphasizes production-grade reliability and hallucination control through grounding, confidence thresholds, validation, retries/fallbacks, and full observability (logging/metrics/traces) with continuous evaluation and feedback loops.”

“ML/GenAI engineer with production experience building an LLM-powered governance layer that summarizes verified drift/performance signals into validation reports and release notes, designed for regulated environments with de-identification and non-blocking fallbacks. Strong Airflow-based orchestration background across healthcare and finance, integrating Databricks/Spark and MLflow for scalable retraining/monitoring. Demonstrated ability to partner with non-technical healthcare operations teams to deliver actionable risk-scoring outputs via dashboards and automated reporting.”

“Built production LLM systems including a real-time customer feedback analysis and workflow automation platform using RAG and multi-agent orchestration with confidence-based human escalation, addressing privacy and legacy integration challenges. Also automated ML operations with Airflow/Kubernetes (e.g., daily churn model retraining) cutting retraining time to under 30 minutes, and demonstrates a rigorous testing/monitoring approach plus strong non-technical stakeholder collaboration.”

“Backend/ML-leaning engineer who built a content-based event recommender for FlowMingle using embeddings + HNSW vector search on Google Cloud, with Firebase as the backend and a managed recommendation lifecycle (15 recs/user, daily async generation, weekly deletion) now serving 1500+ users. Also led a cost-driven migration of ConvAI services to Azure AI using parallel request testing from a Unity client, with post-migration monitoring via logs and model evals; contributed to a Massachusetts law-enforcement conversation analysis system by expanding ingestion to PDF/TXT/Excel and multi-file inputs.”

“AI engineer who has shipped production LLM systems across customer service and marketing use cases—building a RAG app on Azure OpenAI and speeding retrieval with Redis caching tied to Okta sessions. Also implemented a LangGraph multi-agent workflow that pulls image context from Figma to generate structured HTML marketing emails, adding a verification agent to improve image-selection accuracy while optimizing solution cost for business stakeholders.”

“ML/MLOps practitioner who built production systems for telecom network analytics, including an automated labeling + multi-label Random Forest solution that cut labeling effort by 90% and sped up RCA. Led an Ericsson auto-deployment platform using Airflow, Azure IoT Hub, Docker, and Celery to orchestrate 120+ containerized ML/rule-based deployments, saving ~80 hours of setup per deployment.”

“Designed and deployed an enterprise LLM-powered clinical/pharmacy policy knowledge assistant at CVS Health, replacing manual searches across PDFs/Word/SharePoint with a HIPAA-compliant RAG system. Built end-to-end ingestion and orchestration (Airflow + Azure ML/Data Lake + vector index) with PHI masking, versioned re-embedding, and production monitoring (Prometheus/Grafana), and partnered closely with clinicians/compliance to ensure policy-grounded, auditable answers.”

“AI/LLM engineer with recent production experience at UnitedHealth Group building an end-to-end RAG system over structured EMR data and unstructured clinical notes, including evidence retrieval, GPT/LLaMA-based reasoning, and a validation layer for reliability. Strong in orchestration (Kubeflow/Airflow/MLflow), prompt engineering for noisy healthcare text, and rigorous evaluation/monitoring with gold-standard benchmarking, plus close collaboration with clinical operations stakeholders.”

“At Dreamline AI, built and productionized an AWS-based incentive intelligence platform that uses Llama-2/GPT-4 to extract eligibility rules from unstructured state policy documents into structured JSON, then processes them with Glue/PySpark and serves results via Lambda/SageMaker/API Gateway. Designed state-specific ingestion connectors plus schema validation and automated checks/alerts to handle frequent policy/format changes without breaking the pipeline, and partnered with business/analytics stakeholders to deliver interpretable eligibility decisions via explanations and dashboards.”

“Candidate states they have worked on many robotics software system projects and has overcome many technical challenges, but declined to provide any project details during the screening and ended the interview early.”

“ML/LLM engineer at Charles Schwab who built a production loan-advisor chatbot integrated with internal knowledge and loan-calculator APIs, adding strict numeric validation to prevent rate hallucinations and optimizing context to control costs. Also runs ~40 Airflow DAGs orchestrating retraining/ETL/drift monitoring with an automated Snowflake→SageMaker→auto-deploy pipeline, and uses rigorous testing plus canary rollouts tied to business metrics and compliance constraints.”

“Built an AI-driven insurance policy summarization platform at Marsh, taking it end-to-end from messy PDF ingestion/OCR and custom extraction through LLM fine-tuning and AWS SageMaker deployment. Delivered measurable impact (25% reduction in manual review time, 99% uptime) and demonstrated strong production MLOps/LLMOps practices with Airflow/Step Functions orchestration, rigorous evaluation (ROUGE + human review), and continuous monitoring for drift, latency, and hallucinations.”