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

I am an innovative and systems-driven Senior Generative AI Engineer with over 8 years of experience architecting and deploying enterprise-grade AI systems across finance, insurance, and platform engineering. I specialize in the full Generative AI lifecycle including prompt engineering, model tuning, vector retrieval, and observability, leveraging cutting-edge LLM tooling and models such as GPT-4, Claude 3, LLaMA 3, and Mistral. My expertise extends to building secure, scalable, and production-ready AI infrastructures aligned with compliance and performance requirements. I have hands-on experience designing and deploying multi-model orchestration, retrieval-augmented generation pipelines, and multi-agent architectures using tools like LangChain, Hugging Face Transformers, and Kubernetes-managed inference servers. With proficiency in MLOps practices, security governance, and cloud-native deployments, I continuously optimize AI workflows, ensuring robustness, transparency, and scalability in complex enterprise environments.…I am an innovative and systems-driven Senior Generative AI Engineer with over 8 years of experience architecting and deploying enterprise-grade AI systems across finance, insurance, and platform engineering. I specialize in the full Generative AI lifecycle including prompt engineering, model tuning, vector retrieval, and observability, leveraging cutting-edge LLM tooling and models such as GPT-4, Claude 3, LLaMA 3, and Mistral. My expertise extends to building secure, scalable, and production-ready AI infrastructures aligned with compliance and performance requirements. I have hands-on experience designing and deploying multi-model orchestration, retrieval-augmented generation pipelines, and multi-agent architectures using tools like LangChain, Hugging Face Transformers, and Kubernetes-managed inference servers. With proficiency in MLOps practices, security governance, and cloud-native deployments, I continuously optimize AI workflows, ensuring robustness, transparency, and scalability in complex enterprise environments.

Pravalika Sheri

AI Engineer, Data Scientist, AI Developer





I am an innovative and systems-driven Senior Generative AI Engineer with over 8 years of experience architecting and deploying enterprise-grade AI systems across finance, insurance, and platform engineering. I specialize in the full Generative AI lifecycle including prompt engineering, model tuning, vector retrieval, and observability, leveraging cutting-edge LLM tooling and models such as GPT-4, Claude 3, LLaMA 3, and Mistral. My expertise extends to building secure, scalable, and production-ready AI infrastructures aligned with compliance and performance requirements. I have hands-on experience designing and deploying multi-model orchestration, retrieval-augmented generation pipelines, and multi-agent architectures using tools like LangChain, Hugging Face Transformers, and Kubernetes-managed inference servers. With proficiency in MLOps practices, security governance, and cloud-native deployments, I continuously optimize AI workflows, ensuring robustness, transparency, and scalability in complex enterprise environments.…I am an innovative and systems-driven Senior Generative AI Engineer with over 8 years of experience architecting and deploying enterprise-grade AI systems across finance, insurance, and platform engineering. I specialize in the full Generative AI lifecycle including prompt engineering, model tuning, vector retrieval, and observability, leveraging cutting-edge LLM tooling and models such as GPT-4, Claude 3, LLaMA 3, and Mistral. My expertise extends to building secure, scalable, and production-ready AI infrastructures aligned with compliance and performance requirements. I have hands-on experience designing and deploying multi-model orchestration, retrieval-augmented generation pipelines, and multi-agent architectures using tools like LangChain, Hugging Face Transformers, and Kubernetes-managed inference servers. With proficiency in MLOps practices, security governance, and cloud-native deployments, I continuously optimize AI workflows, ensuring robustness, transparency, and scalability in complex enterprise environments.

Available to hire

I am an innovative and systems-driven Senior Generative AI Engineer with over 8 years of experience architecting and deploying enterprise-grade AI systems across finance, insurance, and platform engineering. I specialize in the full Generative AI lifecycle including prompt engineering, model tuning, vector retrieval, and observability, leveraging cutting-edge LLM tooling and models such as GPT-4, Claude 3, LLaMA 3, and Mistral. My expertise extends to building secure, scalable, and production-ready AI infrastructures aligned with compliance and performance requirements.

I have hands-on experience designing and deploying multi-model orchestration, retrieval-augmented generation pipelines, and multi-agent architectures using tools like LangChain, Hugging Face Transformers, and Kubernetes-managed inference servers. With proficiency in MLOps practices, security governance, and cloud-native deployments, I continuously optimize AI workflows, ensuring robustness, transparency, and scalability in complex enterprise environments.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Work Experience

GenAI Engineer at USAA

December 1, 2024 - Present

Architected and implemented retrieval-augmented generation platforms integrating long-form knowledge bases into compliant chat interfaces using LangChain and Qdrant with models like Claude 3 and GPT-4. Fine-tuned open-source LLMs and deployed self-hosted inference APIs ensuring fallback and model sovereignty. Designed multi-model routing systems for intelligent model selection and developed RESTful APIs for AI inference and secure enterprise integration. Built prompt observability and auditing tools, and deployed multi-modal inference workloads for insurance claim triage and content generation on Kubernetes. Orchestrated data and model pipelines via Apache Airflow and TFX, implemented security governance with Vault and IAM, and monitored deployments using Prometheus and Grafana. Developed multi-agent collaborative workflows using LangChain Agents and optimized agent reasoning via reinforcement learning. Established CI/CD pipelines with GitHub Actions and integrated RAG evaluators for m

Senior Machine Learning Engineer at Franklin Templeton

November 30, 2024 - August 13, 2025

Designed scalable machine learning pipelines for daily market data using Airflow and Azure ML Pipelines. Served transformer-based fraud models on Kubernetes clusters with TorchServe and Triton, ensuring low latency through GPU optimization. Built centralized real-time feature stores using Feast and containerized ML workflows with Argo and Helm. Managed model versioning and reproducibility with DVC and MLflow, and implemented explainability via SHAP and LIME for compliance. Deployed infrastructure with Terraform, integrated encrypted secret management and OAuth2 authentication, and automated CI/CD pipelines with GitHub Actions. Developed Conversational Speech LLM-based virtual assistants optimized for domain-specific and edge deployments, and standardized internal ML tooling and deployment templates across teams.

Applied Scientist – NLP/ML at Tansoncorp

February 28, 2023 - August 13, 2025

Built NLP pipelines for customer service ticket normalization and multi-intent classification using spaCy, NLTK, TensorFlow, and Hugging Face Transformers. Developed real-time inference APIs deployed on AWS Lambda for support triage, and automated weekly retraining and validation pipelines with Airflow and Step Functions. Implemented hybrid semantic search combining TF-IDF and transformer embeddings for improved fallback retrieval from PostgreSQL. Processed scanned financial documents with OpenCV and token classification, and developed vector logging for inference auditing. Secured internal APIs with JWT authentication and rate limiting, leveraged AWS Transcribe and Polly for end-to-end conversational AI, and optimized models for real-time edge deployments. Built internal QA tools to visualize embeddings and attention for linguists supporting model validation.

Data Scientist at NIIT Technologies

March 31, 2021 - August 13, 2025

Developed predictive models for customer churn, fraud detection, and pricing using XGBoost, LightGBM, and scikit-learn with cross-validation and hyperparameter tuning. Managed data cleaning, feature engineering, and time-series forecasting with Pandas, NumPy, and Statsmodels. Built lightweight Flask APIs deployed on AWS EC2 for model inference and visualized insights via Seaborn and Plotly. Extracted entities and topics from unstructured feedback using NLP techniques and queried relational databases for analytics. Coordinated version control and deployment best practices, and created dashboards and Excel summaries to communicate model results for stakeholders.

Jr. ML Engineer at Datamatics Global

September 30, 2018 - August 13, 2025

Performed data cleaning and feature preparation for churn prediction models, trained classification algorithms including logistic regression and random forests. Conducted exploratory data analysis using MySQL and PostgreSQL queries and created visualizations with Matplotlib and Seaborn. Evaluated A/B testing results and automated survey data ingestion. Applied text preprocessing for sentiment tagging and keyword extraction and designed Tableau dashboards for churn risk visualization. Maintained version control using Git and produced summary reports for non-technical audiences.