11/03/2026 updated
100 % available
Senior AI / Data Scientist
Lucan, Ireland
Only remote
M.Sc. Data AnalyticsAI AgentsPython (Programming Language)Natural Language ProcessingSQL DatabasesAWS Data AnalyticsAzure Data AnalyticsDatabricks
Agentic AI & LLM Development
Design and implementation of agentic AI systems using LangGraph, including multi-agent workflows for risk detection, site investigation, SQL/data tools and RAG-based retrieval with conditional routing and stateful multi-step reasoning. Integration of guardrails, prompt validation, hallucination mitigation and human-in-the-loop review mechanisms for responsible AI applications.
Machine Learning & Deep Learning
End-to-end development of production-grade machine learning and deep learning models across financial services, healthcare and life sciences domains. Experience includes credit-risk prediction pipelines, anomaly detection with autoencoders, CNN-based retail planogram optimization, time-series forecasting and NLP-driven ICD-10 coding solutions.
Data Engineering & MLOps
Building and automating data pipelines using Airflow, Snowflake, Azure Databricks and PySpark. Expertise in MLflow tracking, model registry, CI/CD automation, self-healing ETL pipelines and LLM/RAG evaluation frameworks for regression testing before production releases.
Cloud Platforms (Azure, AWS, GCP)
Hands-on experience deploying and managing machine learning and analytics solutions on Microsoft Azure, AWS and Google Cloud Platform, including Azure AI Search, Azure Databricks and related cloud-native services.
Natural Language Processing (NLP)
Application of NLP techniques for clinical note analysis, ICD-10 code recommendation, chatbot development using LLaMA-3 and RAG, and retrieval-augmented generation systems for operational insights.
Python & FastAPI
Extensive use of Python for data science, machine learning and API development. Exposure of agent capabilities through FastAPI APIs for clinical and operational stakeholders.
Statistical Risk Monitoring
Application of statistical risk-monitoring techniques to analyze site-level KPIs and KRIs, identify deviations and emerging trends, and compare site performance against study-level baselines in regulated clinical environments.
Predictive & Causal Modelling
Delivery of predictive and causal models to guide product decisions, optimize feature adoption and support pricing and budgeting decisions across telecom, healthcare and retail domains.
Docker & CI/CD
Containerization of applications using Docker and implementation of CI/CD pipelines to ensure reliable, automated deployment and monitoring of machine learning models and data platforms.
Design and implementation of agentic AI systems using LangGraph, including multi-agent workflows for risk detection, site investigation, SQL/data tools and RAG-based retrieval with conditional routing and stateful multi-step reasoning. Integration of guardrails, prompt validation, hallucination mitigation and human-in-the-loop review mechanisms for responsible AI applications.
Machine Learning & Deep Learning
End-to-end development of production-grade machine learning and deep learning models across financial services, healthcare and life sciences domains. Experience includes credit-risk prediction pipelines, anomaly detection with autoencoders, CNN-based retail planogram optimization, time-series forecasting and NLP-driven ICD-10 coding solutions.
Data Engineering & MLOps
Building and automating data pipelines using Airflow, Snowflake, Azure Databricks and PySpark. Expertise in MLflow tracking, model registry, CI/CD automation, self-healing ETL pipelines and LLM/RAG evaluation frameworks for regression testing before production releases.
Cloud Platforms (Azure, AWS, GCP)
Hands-on experience deploying and managing machine learning and analytics solutions on Microsoft Azure, AWS and Google Cloud Platform, including Azure AI Search, Azure Databricks and related cloud-native services.
Natural Language Processing (NLP)
Application of NLP techniques for clinical note analysis, ICD-10 code recommendation, chatbot development using LLaMA-3 and RAG, and retrieval-augmented generation systems for operational insights.
Python & FastAPI
Extensive use of Python for data science, machine learning and API development. Exposure of agent capabilities through FastAPI APIs for clinical and operational stakeholders.
Statistical Risk Monitoring
Application of statistical risk-monitoring techniques to analyze site-level KPIs and KRIs, identify deviations and emerging trends, and compare site performance against study-level baselines in regulated clinical environments.
Predictive & Causal Modelling
Delivery of predictive and causal models to guide product decisions, optimize feature adoption and support pricing and budgeting decisions across telecom, healthcare and retail domains.
Docker & CI/CD
Containerization of applications using Docker and implementation of CI/CD pipelines to ensure reliable, automated deployment and monitoring of machine learning models and data platforms.
Experience
Professional experience5 - 10 years
Freelancer since2025
Languages
EnglishFluent
Project history
Built an agentic AI assistant for clinical-trial risk monitoring (RMCM/SPT), enabling site managers and central monitors to investigate site performance, retrieve KPIs/KRIs, identify emerging risk signals and generate evidence-backed operational insights. Designed multi-agent workflows using LangGraph, orchestrating risk detection, site investigation, SQL/data tools and RAG-based retrieval. Applied statistical risk-monitoring techniques and implemented human-in-the-loop controls, guardrails, structured outputs, tracing and LLM/agent evaluation. Exposed agent capabilities through Python/FastAPI APIs.
Integrated Responsible AI practices into LLM applications by implementing guardrails, prompt validation, hallucination mitigation and human-in-the-loop review mechanisms. Built and deployed an end-to-end credit-risk prediction pipeline on Azure Databricks covering feature engineering, model training, MLflow tracking, model registry and CI/CD automation. Developed an AI-driven self-healing ETL pipeline and an automated LLM and RAG evaluation framework to measure retrieval relevance, groundedness, response quality and hallucination rates.
Modeled product impact on customer success metrics and partnered with product/clinical teams on data quality and prioritization. Delivered predictive and causal models to guide product decisions and optimize feature adoption. Automated data pipelines in Airflow and Snowflake, cutting reporting time by 50%.