Available to hire
Hi, I’m Sneh Pankajbhai Vora, an AI/ML Engineer with a passion for turning data into practical, scalable solutions. I design and deploy GenAI assistants, retrieval-based workflows, and fraud detection/risk analysis platforms that improve decision-making and operational efficiency.
I collaborate with cross-functional teams in secure, cloud-based environments from data preparation to validation and deployment, prioritizing robust prompts, guardrails, HIPAA-aligned usage, and observability to deliver reliable AI at scale.
Skills
Work Experience
AI/ML Engineer – Gen AI at Molina Healthcare USA
January 1, 2025 - PresentLed GenAI initiatives including a RAG-based assistant using Python and LangChain, leveraging Azure OpenAI and Azure AI Search for document retrieval to support healthcare policy documents and authorization guidelines, reducing manual lookup time by 30%. Built document ingestion pipelines with PyMuPDF to process 8,000+ pages of policies, SOPs, and authorization documents; stored assets in Azure Blob Storage and indexed chunks in Azure AI Search for scalable semantic retrieval. Implemented embedding-based retrieval using FAISS with metadata filters to improve retrieval accuracy by 22%. Created LLM-based Summarization prompts for authorization notes, claim reviews, denial reasons, and policy references, reducing case-review prep by 20–25%. Integrated GenAI workflows with healthcare systems via SQL and REST APIs. Applied prompt engineering, few-shot prompting, guardrails, PHI masking, and RBAC to ensure HIPAA-aligned usage. Assisted in LLM evaluation with Python, pandas, and MLflow acros
Machine Learning Engineer at Accenture India
June 1, 2021 - November 1, 2023Developed an ML-based fraud detection and risk scoring platform for a confidential BFSI client, analyzing 1M+ transactions to identify suspicious activity and fraud patterns. Built preprocessing and feature engineering pipelines using Python, Pandas, NumPy, and SQL to prepare data for model training and batch scoring. Engineered 40+ fraud-risk features (e.g., transaction frequency, amount deviation, merchant category, device behavior, location mismatch, account age, payment channel, historical indicators). Trained and validated an XGBoost classifier; implemented Isolation Forest for anomaly detection. Performed hyperparameter tuning, cross-validation, and threshold optimization, improving model performance by 18–22% over baselines. Addressed class imbalance with SMOTE, undersampling, and class-weight tuning, boosting recall by ~20% while controlling false positives. Evaluated models with Precision, Recall, F1, ROC-AUC (~0.86), and PR curves. Automated batch scoring and risk assessmen
Education
Master of Science in Computer Science at New Jersey Institute of Technology
January 11, 2030 - June 29, 2026Bachelor of Technology in Computer Science & Engineering at Charotar University of Science and Technology
January 11, 2030 - June 29, 2026Qualifications
IBM Certified Data Architect - Big Data
January 11, 2030 - June 29, 2026Machine Learning Course - Wissenaire, IIT Bhubaneswar
January 11, 2030 - June 29, 2026Industry Experience
Healthcare, Financial Services, Professional Services, Software & Internet
Skills
Hire a AI Engineer
We have the best ai engineer experts on Twine. Hire a ai engineer today.