Prannay Khushalani
Open to AI & Data Engineering Roles

Prannay
Khushalani

AI Engineer & Data Engineer

I build production AI systems and scalable data pipelines. From RAG architectures and LLM applications to Spark-powered data warehouses and end-to-end ML pipelines.

RAG & LLM Systems Data Pipelines ML Engineering Vector Search Snowflake · Spark MLOps
4 Peer-Reviewed Publications IEEE · Springer · AIP
9 Projects Shipped AI · Data · ML
85% Chatbot Accuracy Lift Owlie RAG System
88.4% Default Rate Reduction Credit Risk Model

About Me

I build at the intersection of AI and data, shipping intelligent systems that are grounded in robust, scalable infrastructure.

🤖

AI & LLM Engineering

I design and ship production RAG pipelines, LLM-powered applications, and vector search systems. Built Owlie, a university chatbot indexing 5,000+ web pages with LLAMA3-70B, FAISS, and Sentence Transformers. Achieved an 85% accuracy boost with sub-3 second response times.

⚙️

Data Engineering

I build scalable ETL/ELT pipelines, data warehouses, and analytics infrastructure. At Recykal, I processed 1.2M+ records using SQL and Python, built KPI pipelines that cut reporting time by 20%, and designed Power BI dashboards used across 20+ Indian states.

📊

ML Research & Engineering

I handle the full model lifecycle, from feature engineering to deployment. Reduced credit loan default rates from 25.8% to 3% using XGBoost. Published 4 peer-reviewed papers in IEEE, Springer, and AIP on NLP and predictive modeling. MS in Business Analytics and AI from UT Dallas (GPA 3.45).

What I Build

🤖

AI & LLM Engineering

I build production RAG systems with custom vector indexing, LLM orchestration via LangChain and LlamaIndex, fine-tuning with LoRA and PEFT, and AI-powered web applications deployed on Vercel and Render.

RAG Pipelines LangChain FAISS · Pinecone Prompt Engineering Fine-tuning
⚙️

Data Engineering

I design scalable ETL and ELT pipelines with PySpark and Databricks, set up data warehouses on Snowflake, orchestrate workflows with Airflow and dbt, and build analytics dashboards in Power BI and Tableau.

PySpark Snowflake Databricks Airflow · dbt ETL / ELT
📊

ML Engineering

I handle the full model lifecycle: EDA, feature engineering, training with XGBoost, PyTorch, and scikit-learn, experiment tracking with MLflow and W&B, and containerized model deployment via Docker and FastAPI.

XGBoost · PyTorch MLflow FastAPI Docker Model Deployment

Tech Stack

AI Engineering Stack
Data Engineering Stack
ML & Cloud Stack

Core Concepts

Retrieval-Augmented Generation LLM Fine-tuning (LoRA/PEFT) Prompt Engineering ETL / ELT Pipelines Data Warehousing Vector Search Feature Engineering Ensemble Methods Transformers MLOps & CI/CD Stream Processing Zero/Few-shot Learning Time Series Analysis A/B Testing Speech Recognition (ASR)

Explore More

🚀

Projects & Publications

RAG systems, ML models, data pipelines, and 4 peer-reviewed papers in IEEE, Springer & AIP.

💼

Education & Experience

MS in Business Analytics & AI from UT Dallas, industry experience, certifications and awards.

📬

Get In Touch

Open to AI & Data Engineering roles. Let's connect and build something impactful together.