@yugesh_verma - Data Scientist

Data Scientist

AI-native engineer

Available for hire

Years of experience

1+ years

Experience level

Mid-level

Available for

Full-time, Contract

Available from

30 Jul 2026

Download Resume / CV

Data Scientist and ML Engineer with 5+ years of experience building production AI systems across machine learning, NLP, generative AI, and computer vision. Currently at Artivatic.ai where I own end-to-end ML pipelines — from feature engineering and model training through deployment, monitoring, and retraining. Key achievements include automating 70% of life insurance underwriting decisions and reducing LLM inference costs by 90% through a confidence-based routing architecture. I work with Python, XGBoost, LightGBM, LangChain, LangGraph, BERT, RAG pipelines, OpenAI API, and AWS in production. Looking for remote Data Scientist, ML Engineer, or AI/GenAI Engineer roles where I can apply this depth across diverse problem domains and contribute to a team building meaningful AI-driven products.

Employment History

data scientist at artivatic data labs Current 2024 - Now

At Artivatic.ai I work as a Data Scientist responsible for building and operating production AI systems for life and health insurance clients. My current responsibilities include: Developing and maintaining a family of product-specific ML models for life insurance underwriting across Term, Savings & Investment, and Minor Life plans — built using XGBoost, LightGBM, and CatBoost with full SHAP-based explainability. These models currently automate 70% of application decisions at production scale with zero human involvement. Managing the complete model lifecycle including quarterly retraining cycles to track current trends, annual feature review and version management, and production monitoring using Evidently for drift detection and Metabase dashboards for performance visibility. Building and operating a healthcare claims adjudication engine that classifies medical bill line items as payable or non-payable using a 4-layer cascade pipeline — exact match lookup, BERT-based semantic cluster search, fuzzy matching, and GPT-4 fallback — reducing LLM inference cost by 90% in production. Architecting a multi-agent AI underwriting worksheet system using LangChain, LangGraph, and OpenAI API where four specialised domain agents interpret model outputs and a central LLM orchestration agent synthesises them into a structured report for human underwriters. Building an end-to-end document intelligence pipeline handling classification across 64 document categories, key-value extraction, and Aadhaar compliance masking — deployed on AWS with Docker, Kubernetes, and FastAPI. Collaborating with actuarial SMEs and business stakeholders to translate domain knowledge into ML features and explainable AI systems that are trusted and adopted in production.

data scientist at data science wizards 2022 - 2023

At Data Science Wizards I worked as a Data Scientist responsible for building production ML models and leading proof-of-concept development across multiple insurance and industrial domains. Designed and deployed a health insurance customer persistency and churn prediction model using XGBoost in R and Python against MSSQL data — enabling insurers to identify at-risk policyholders early and design proactive retention strategies before lapse occurs. Led proof-of-concept projects across multiple domains including visual uniqueness detection for individual animal identification in livestock insurance using MobileNet and TensorFlow Lite, time-series production performance and maintenance window prediction for industrial printing press machinery using XGBoost and Prophet, and long-form medical research paper summarisation using Google PaLM for healthcare professionals. Contributed to UnifyAI, a production-grade MLOps platform built as an internal product, as technology research lead — evaluating and selecting the full toolchain including Feast for feature store management, MLflow for experiment tracking, Seldon Core for model serving, Kubernetes for orchestration, Evidently for monitoring, and Grafana for observability dashboards. Collaborated with cross-functional teams on problem scoping, model development, and stakeholder presentations across insurance, healthcare, and industrial domains.

data scientist at analytics india magazine 2021 - 2021

At Analytics India Magazine I worked as a Data Science Content Writer responsible for researching, authoring, and publishing technical blogs and tutorials for one of India's largest data science communities. Researched and wrote in-depth technical articles and hands-on tutorials covering machine learning algorithms, deep learning, neural networks, statistical modelling, and practical AI tool usage — helping practitioners understand and apply complex concepts through clear, example-driven writing. Developed educational content and step-by-step tutorials using Python, Scikit-learn, TensorFlow, PyTorch, Pandas, Flask, and REST APIs — covering topics from model building basics to production deployment patterns. Collaborated with the editorial team to maintain technical accuracy, relevance, and clarity across all published content — articles published at analyticsindiamag.com. Built a deep conceptual foundation in ML and AI through the discipline of explaining complex topics clearly — the understanding gained through technical writing directly shaped the production engineering work that followed in later roles.

Education

bachelor of engineering at rajiv gandhi technical university 2013 - 2017

Data Scientist

Skills

Languages

Employment History

Education

Get realtime job alerts