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

I'm Smriti Karanjit, an AI/ML Engineer focused on NLP, generative AI, and MLOps. I design and deploy production-grade ML systems in financial services and enterprise settings, building real-time pipelines, LLM-driven extraction, and scalable model lifecycle on AWS and Azure. In my current role at State Street, I process 500K+ daily financial data feeds, optimize fraud detection to 98.7% precision, and cut analyst research time by 60%. I translate unstructured data into risk signals to support risk, compliance, and operations teams.…I'm Smriti Karanjit, an AI/ML Engineer focused on NLP, generative AI, and MLOps. I design and deploy production-grade ML systems in financial services and enterprise settings, building real-time pipelines, LLM-driven extraction, and scalable model lifecycle on AWS and Azure. In my current role at State Street, I process 500K+ daily financial data feeds, optimize fraud detection to 98.7% precision, and cut analyst research time by 60%. I translate unstructured data into risk signals to support risk, compliance, and operations teams.

Smriti Karanjit

AI Engineer, AI Developer, Data Scientist, +2





I'm Smriti Karanjit, an AI/ML Engineer focused on NLP, generative AI, and MLOps. I design and deploy production-grade ML systems in financial services and enterprise settings, building real-time pipelines, LLM-driven extraction, and scalable model lifecycle on AWS and Azure. In my current role at State Street, I process 500K+ daily financial data feeds, optimize fraud detection to 98.7% precision, and cut analyst research time by 60%. I translate unstructured data into risk signals to support risk, compliance, and operations teams.…I'm Smriti Karanjit, an AI/ML Engineer focused on NLP, generative AI, and MLOps. I design and deploy production-grade ML systems in financial services and enterprise settings, building real-time pipelines, LLM-driven extraction, and scalable model lifecycle on AWS and Azure. In my current role at State Street, I process 500K+ daily financial data feeds, optimize fraud detection to 98.7% precision, and cut analyst research time by 60%. I translate unstructured data into risk signals to support risk, compliance, and operations teams.

Available to hire

I’m Smriti Karanjit, an AI/ML Engineer focused on NLP, generative AI, and MLOps. I design and deploy production-grade ML systems in financial services and enterprise settings, building real-time pipelines, LLM-driven extraction, and scalable model lifecycle on AWS and Azure.

In my current role at State Street, I process 500K+ daily financial data feeds, optimize fraud detection to 98.7% precision, and cut analyst research time by 60%. I translate unstructured data into risk signals to support risk, compliance, and operations teams.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Language

English

Fluent

Work Experience

AI/ML Engineer at State Street Corporation

August 1, 2025 - Present

Architected a real-time NLP pipeline (Python, Kafka, BERT) processing 500K+ daily financial data feeds to generate live risk signals for portfolio and compliance teams. Led development of LLM-driven extraction pipelines (LangChain + GPT) to convert earnings calls and SEC filings into structured insights, reducing analyst research time by 60% and enabling reporting across 200+ documents per quarter. Built a scalable ML feature store with automated retraining on AWS SageMaker and drift detection (Evidently AI), preserving performance within 2% of baseline. Implemented fraud detection using Isolation Forest and LSTM achieving 98.7% precision, cutting compliance workload by 35%. Established CI/CD style evaluation with MLflow and scikit-learn, standardized production model selection, and automated data ingestion/serving with Airflow, Docker, and AWS ECS, maintaining 99.9% uptime. Created Power BI and Streamlit dashboards to translate model outputs into risk metrics for stakeholders across r

Machine Learning Engineer at Kellton Tech Solutions Ltd.

July 1, 2021 - December 1, 2023

Developed ML models for churn prediction and recommendation across e-commerce and insurance clients, improving customer retention by 20%. Built a real-time collaborative filtering engine using matrix factorization (Surprise) across 500K+ monthly users. Applied NLP techniques (TF-IDF, Word2Vec, fine-tuned BERT) to analyze 1M+ customer feedback records. Built and optimized ETL pipelines (SQL, Python, PySpark) processing 10M+ records, improving training efficiency by 35%. Implemented automated model monitoring and retraining via Apache Airflow, enabling proactive drift detection and reducing model degradation incidents by 50%. Partnered with business teams to deliver Power BI dashboards translating ML outputs into actionable churn risk insights.