building RAG systems
I build and optimize production RAG & search systems for enterprise — vector databases, LLM inference, and machine learning that survives outside the notebook.
Hey! 😄 My name is Shaun Parimoo.
I'm an AI/ML Engineer from Warren, NJ, currently based in Philadelphia, PA. I studied Biological Sciences with a Chemistry minor at the University of Pittsburgh, then earned my master's in Computational Biomedicine & Biotechnology at University of Pittsburgh School of Medicine.
At my core I'm a problem solver. My roots are in computational biology and data science, building predictive models for drug discovery, classification systems, and statistical modeling. The tools change but the drive is the same: break down complex problems, find the signal in the noise, and build systems that deliver real insights.
Today, I build and optimize production Agentic and RAG search systems that serve enterprise clients at scale. My work spans the full AI stack, from vector database architecture with Weaviate and Elasticsearch, to LLM inference optimization focused on throughput and search quality, to model lifecycle management including deployment, monitoring, and model output quality. I care about making AI systems that don't just work in a notebook, but survive and thrive in production.
In my free time I'm an avid Soccer player and fan (Lets Go Brighton!) ⚽.
The stack I use to build and ship production AI systems
Python, SQL, JavaScript, and Shell scripting for building backend services and ML pipelines. FastAPI for high-performance REST APIs; GraphQL for flexible data access.
Python
JavaScript
FastAPI
GraphQL
MySQL
R
Bash
Building RAG pipelines, vector search, and embedding systems. vLLM for inference; Weaviate, Elasticsearch, and OpenSearch for retrieval; MLflow for experiment tracking.
PyTorch
vLLM
Weaviate
Elasticsearch
OpenSearch
MLflow
Deploying and scaling ML systems with Kubernetes, Docker, and Helm across Azure, AWS, and GCP. CI/CD with CircleCI, load testing with Locust, monitoring with OpenSearch dashboards. Redis/KeyDB for caching.
Kubernetes
Docker
Helm
CircleCI
Git
Redis
Azure
AWS
GCP
Certified in Azure Fundamentals (AZ-900) and Supervised Machine Learning (DeepLearning.AI). Continual investment in cloud and ML depth.
Machine learning for drug discovery — predicting effective combination therapies with LASSO feature selection, PCA, and GridSearchCV-tuned models evaluated by AUC-ROC.
Python
scikit-learn
LASSO
PCA
GridSearchCV
AUC-ROC
A data-science approach to early Parkinson's detection, comparing XGBoost and Random Forest classifiers with PCA dimensionality reduction and GridSearchCV tuning.
Python
XGBoost
Random Forest
PCA
GridSearchCV
I'm always looking to collaborate on AI/ML engineering, RAG systems, and LLM infrastructure projects! Whether you need expertise in search and retrieval optimization, vector database architecture, or building production AI systems, I'd love to discuss how my skills can help your organization.
Based in Philadelphia and available for remote consulting engagements. Reach out via email to discuss potential collaborations, consulting opportunities, or if you have questions about my projects.