Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I’m Alexis Rodriguez, a Senior AI/ML Engineer with 10+ years building and shipping production-grade AI systems across healthcare and financial services. I specialize in agentic AI and hybrid RAG, and I’ve led end-to-end deployments that improve real-world outcomes—such as reducing clinician documentation time by 30–40% while preserving HIPAA-compliant auditability. I’m especially focused on LLMOps: evaluation, observability, governance, and safe continuous deployment. From multi-agent orchestration (A2A) and MCP-style tool authorization to low-latency inference optimization and drift monitoring, I aim to reduce latency and hallucinations while accelerating delivery through measurable, empirically grounded engineering.…I’m Alexis Rodriguez, a Senior AI/ML Engineer with 10+ years building and shipping production-grade AI systems across healthcare and financial services. I specialize in agentic AI and hybrid RAG, and I’ve led end-to-end deployments that improve real-world outcomes—such as reducing clinician documentation time by 30–40% while preserving HIPAA-compliant auditability. I’m especially focused on LLMOps: evaluation, observability, governance, and safe continuous deployment. From multi-agent orchestration (A2A) and MCP-style tool authorization to low-latency inference optimization and drift monitoring, I aim to reduce latency and hallucinations while accelerating delivery through measurable, empirically grounded engineering.

Alexis Rodriguez

AI Engineer, Data Scientist, Software Engineer, +1





I’m Alexis Rodriguez, a Senior AI/ML Engineer with 10+ years building and shipping production-grade AI systems across healthcare and financial services. I specialize in agentic AI and hybrid RAG, and I’ve led end-to-end deployments that improve real-world outcomes—such as reducing clinician documentation time by 30–40% while preserving HIPAA-compliant auditability. I’m especially focused on LLMOps: evaluation, observability, governance, and safe continuous deployment. From multi-agent orchestration (A2A) and MCP-style tool authorization to low-latency inference optimization and drift monitoring, I aim to reduce latency and hallucinations while accelerating delivery through measurable, empirically grounded engineering.…I’m Alexis Rodriguez, a Senior AI/ML Engineer with 10+ years building and shipping production-grade AI systems across healthcare and financial services. I specialize in agentic AI and hybrid RAG, and I’ve led end-to-end deployments that improve real-world outcomes—such as reducing clinician documentation time by 30–40% while preserving HIPAA-compliant auditability. I’m especially focused on LLMOps: evaluation, observability, governance, and safe continuous deployment. From multi-agent orchestration (A2A) and MCP-style tool authorization to low-latency inference optimization and drift monitoring, I aim to reduce latency and hallucinations while accelerating delivery through measurable, empirically grounded engineering.

Available to hire

I’m Alexis Rodriguez, a Senior AI/ML Engineer with 10+ years building and shipping production-grade AI systems across healthcare and financial services. I specialize in agentic AI and hybrid RAG, and I’ve led end-to-end deployments that improve real-world outcomes—such as reducing clinician documentation time by 30–40% while preserving HIPAA-compliant auditability.

I’m especially focused on LLMOps: evaluation, observability, governance, and safe continuous deployment. From multi-agent orchestration (A2A) and MCP-style tool authorization to low-latency inference optimization and drift monitoring, I aim to reduce latency and hallucinations while accelerating delivery through measurable, empirically grounded engineering.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Language

English

Advanced

Work Experience

Staff ML Engineer, Applied AI at HCA Healthcare

September 1, 2025 - May 1, 2026

Led end-to-end architecture and production deployment of a multi-agent hybrid RAG platform for clinical workflows. Integrated EHR context, enterprise knowledge graphs, and vector retrieval to reduce clinician documentation time by 30–40% while maintaining HIPAA-compliant auditability. Designed an agent-to-agent orchestration protocol with a lightweight MCP layer for standardized state handoffs, tool authorization, and encrypted context passing across voice capture, STT, summarization, and EHR-intent agents. Built GCP-native inference pipelines (Vertex AI + Anthos) with intelligent on-prem/cloud routing to support low-latency autoscaling for PHI-sensitive workloads under HIPAA and NIST AI RMF guardrails. Implemented a hybrid retrieval layer (BM25 + dense + structured metadata + graph-linked longitudinal records) to improve relevance and reduce hallucinations. Established an agent evaluation framework (A/B testing, model evals, behavioral drift detection, automated rollback triggers) a

Senior AI/ML Engineer at Andor Health

May 1, 2022 - August 1, 2025

Architected a unified, domain-adapted multi-agent LLM framework for text and voice AI (STT → LLM → TTS), using ReAct-style reasoning, structured tool calling, and MCP-based I/O contracts for composable and auditable agent behaviors. Implemented Pinecone-based semantic retrieval and GraphRAG with Neo4j over clinical knowledge graphs, improving answer accuracy by 40% and enabling multi-hop clinical reasoning. Designed and deployed PEFT pipelines (QLoRA, adapters, instruction tuning) on GPT-4 and Llama 3 via LangChain/LangGraph, reducing fine-tuning time by 40% and cutting hallucinations in clinical workflows by 60%. Built AWS production MLOps using Terraform for EKS/ECR and serverless orchestration (Lambda, Step Functions, API Gateway) to scale throughput 3× with sub-200ms p99 latency. Developed continuous LLM evaluation (LangSmith + RAGAS) and continuous RLHF pipelines (PPO/GRPO + DPO) with Arize AI for behavioral monitoring and drift detection. Optimized inference with vLLM/Tensor

Senior Machine Learning Engineer at Hatch AI

July 1, 2019 - May 1, 2022

Built advanced NLP pipelines for financial document processing on Databricks and Snowflake, including BERT-based NER with spaCy/transformers, reducing extraction error rates from 11% to 2% at high volume. Developed multi-agent conversational systems orchestrated with GPT-3 (pre-dating modern frameworks) using Haystack and custom scaffolding, cutting human intervention by 60% and accelerating analyst workflows. Created an early RLHF feedback loop on Vertex AI using Weights & Biases; expert feedback integration reduced false positives in contract clause detection by 25%. Implemented a real-time LLM summarization platform for financial news, reducing analyst research time by 75% and improving time-to-insight. Built a sentiment analysis engine for investment signals with PyTorch and GCP AutoML, reaching 92% accuracy across 500k+ articles to support data-driven trading decisions. Orchestrated end-to-end MLOps with Kubeflow and Terraform, improving deployment cycles by 30% with automated reg

ML Developer at IBM

February 1, 2016 - June 1, 2019

Engineered an automated OCR and document intelligence pipeline (Tesseract, OpenCV, Keras) for client onboarding, reducing manual data entry from 100,000 to 30,000 hours annually and enabling enterprise document automation at scale. Increased product engagement by 45% through a personalization recommendation engine using scikit-learn and PyTorch. Built a churn prediction system (Logistic Regression, Random Forest, Gradient Boosting) achieving 89% recall on churners and reducing monthly churn by 12% via targeted campaigns. Developed clustering pipelines (K-Means, DBSCAN, hierarchical clustering) for customer segmentation, improving offer redemption by 17%. Deployed a healthcare document classification system (TensorFlow, NLTK) with 97% accuracy across 1,000+ daily records, reducing manual triage overhead. Built time-series forecasting models (ARIMA, LSTM, GRU) for risk and trend prediction to support proactive planning and early anomaly identification. Created executive-grade Tableau das