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

I am Ayaan Ahmed, a Senior AI Engineer with over 10 years of experience applying GenAI and machine learning across banking, healthcare, government, legal, telecom, and customer analytics. I focus on practical decision support rather than just research prototypes, delivering solutions that engineers and frontline teams can actually adopt. I work hands-on with LangGraph, LangChain, LlamaIndex, CrewAI, and RAG to build multiagent assistants that ground answers in real loan files, clinical notes, policy documents, and operational data. I design GenAI-enabled underwriting and risk-scoring flows that combine LLM reasoning with structured models, improve narratives and evidence trails, and deploy scalable data and microservice foundations on AWS and Azure so intelligent workloads run reliably alongside existing systems.…I am Ayaan Ahmed, a Senior AI Engineer with over 10 years of experience applying GenAI and machine learning across banking, healthcare, government, legal, telecom, and customer analytics. I focus on practical decision support rather than just research prototypes, delivering solutions that engineers and frontline teams can actually adopt. I work hands-on with LangGraph, LangChain, LlamaIndex, CrewAI, and RAG to build multiagent assistants that ground answers in real loan files, clinical notes, policy documents, and operational data. I design GenAI-enabled underwriting and risk-scoring flows that combine LLM reasoning with structured models, improve narratives and evidence trails, and deploy scalable data and microservice foundations on AWS and Azure so intelligent workloads run reliably alongside existing systems.

Ayaan Ahmed

AI Engineer, Data Scientist, Full Stack Developer, +2





I am Ayaan Ahmed, a Senior AI Engineer with over 10 years of experience applying GenAI and machine learning across banking, healthcare, government, legal, telecom, and customer analytics. I focus on practical decision support rather than just research prototypes, delivering solutions that engineers and frontline teams can actually adopt. I work hands-on with LangGraph, LangChain, LlamaIndex, CrewAI, and RAG to build multiagent assistants that ground answers in real loan files, clinical notes, policy documents, and operational data. I design GenAI-enabled underwriting and risk-scoring flows that combine LLM reasoning with structured models, improve narratives and evidence trails, and deploy scalable data and microservice foundations on AWS and Azure so intelligent workloads run reliably alongside existing systems.…I am Ayaan Ahmed, a Senior AI Engineer with over 10 years of experience applying GenAI and machine learning across banking, healthcare, government, legal, telecom, and customer analytics. I focus on practical decision support rather than just research prototypes, delivering solutions that engineers and frontline teams can actually adopt. I work hands-on with LangGraph, LangChain, LlamaIndex, CrewAI, and RAG to build multiagent assistants that ground answers in real loan files, clinical notes, policy documents, and operational data. I design GenAI-enabled underwriting and risk-scoring flows that combine LLM reasoning with structured models, improve narratives and evidence trails, and deploy scalable data and microservice foundations on AWS and Azure so intelligent workloads run reliably alongside existing systems.

Available to hire

I am Ayaan Ahmed, a Senior AI Engineer with over 10 years of experience applying GenAI and machine learning across banking, healthcare, government, legal, telecom, and customer analytics. I focus on practical decision support rather than just research prototypes, delivering solutions that engineers and frontline teams can actually adopt.

I work hands-on with LangGraph, LangChain, LlamaIndex, CrewAI, and RAG to build multiagent assistants that ground answers in real loan files, clinical notes, policy documents, and operational data. I design GenAI-enabled underwriting and risk-scoring flows that combine LLM reasoning with structured models, improve narratives and evidence trails, and deploy scalable data and microservice foundations on AWS and Azure so intelligent workloads run reliably alongside existing systems.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Work Experience

Senior AI Engineer/Gen AI at BHG Financial

January 1, 2025 - Present

Led the development of an Intelligent Loan Underwriting Assistant using LangGraph, AWS Bedrock, and RAG pipelines, reducing manual underwriting effort and boosting throughput across a distributed bank network. Architected an event-driven microservices stack on AWS EKS with Kafka to enable real-time underwriting workflows and autonomous agents. Built multi-source ingestion pipelines from loan apps, credit bureau data, and transactions into a unified data layer; standardized borrower data with AWS Glue; stored curated data zones in S3 and low-latency lookups in DynamoDB. Grounded LLM outputs with borrower data using Pinecone and Qdrant, tuned prompts with Chain-of-Thought and ReAct, and conducted A/B RAG retrieval experiments with MLflow. Containerized services, deployed on EKS, automated CI/CD with GitHub Actions, and provisioned infrastructure with Terraform. Implemented monitoring and drift checks with CloudWatch, LangSmith, and Evidently AI; wrote comprehensive tests with PyTest and

AI/ML Engineer at Optum

February 1, 2023 - December 1, 2024

Delivered end-to-end data science for payment integrity and population health; set up the Azure analytics stack using Databricks, Data Factory, Airflow, ADLS Gen2, and Snowflake to support batch retraining and near realtime fraud checks. Built ingestion pipelines bringing claims, pharmacy transactions, enrollment, and prior authorizations into consistent training datasets; created clinical features from claims data; applied Great Expectations to ensure data quality; maintained versioned feature stores; indexed ClinicalBERT note embeddings in FAISS for reuse by ML and the document intelligence layer. Finetuned ClinicalBERT, T5, and GPT-4 via Azure OpenAI for summarization and Q&A, combining tree-based models with LangChain retrieval workflows. Deployed scalable services with AKS, autoscaling, and CI/CD pipelines with GitHub Actions; monitored performance with Prometheus, Grafana, and Azure Monitor.

Data Scientist & Python Developer at State of Minnesota

June 1, 2021 - December 1, 2022

Worked on MN Health Bridge, a population-health risk stratification platform, building a cloud-native Lakehouse on Azure Databricks and ADLS Gen2 to integrate Medicaid claims, HL7 FHIR data, and Social Determinants of Health into a unified feature store for predictive modeling and BI reporting. Built modular ETL pipelines, created clinical features, and implemented data quality checks with Great Expectations. Maintained versioned feature stores, exposed model-ready tables, and built FastAPI endpoints for scoring services; collaborated with DHS staff to interpret results and adjust thresholds.

Data Scientist at Thomson Reuters

January 1, 2020 - May 1, 2021

Explored invoice and matter data to identify spending patterns and labeling gaps, built models predicting potential billing noncompliance, and used TFIDF with n-grams for narrative-driven features. Trained and evaluated models with cross-validation and metrics such as precision, recall, F1, and ROC-AUC; packaged training and batch scoring for AWS-based workflows and integrated predictions into downstream dashboards.

Data Engineer at Motorola Solutions

April 1, 2018 - December 1, 2019

Designed a cloud data lake forPublic safety telemetry and incident data; developed batch and near real-time ETL in Spark and Airflow, standardizing schemas across jurisdictions and loading into Redshift/BigQuery. Transformed raw telemetry into ML-ready feature tables, prototyped ML models for incident forecasting and response time, and productionized best models via Airflow-driven pipelines. Implemented data quality checks, tuned Spark jobs, and documented end-to-end pipelines for knowledge transfer.

Data Engineer at Value Labs

June 1, 2015 - January 1, 2018

Built a Hadoop and Spark-based customer analytics platform to support churn prediction and upsell strategies. Developed ETL jobs over Spark/Hive, created feature-ready tables, and automated data ingestion from on-premises sources via Sqoop. Implemented reproducible ML prep utilities, designed star/snowflake schemas for BI and ML workloads, and orchestrated pipelines with cron/early schedulers.