Sowmya Laveti profile photo 1 on Tal

Sowmya Laveti

AI/ML Engineer

Cleveland

7+ years of work experience

Lucid Motors
Aug 202510 mosPresent

Newark, California, United States

Architecting enterprise Generative AI systems integrating Retrieval-Augmented Generation (RAG) pipelines for internal knowledge and diagnostics platforms. • Designing embedding pipelines using transformer-based models and implementing FAISS vector indexing for se- mantic retrieval. • Implementing hybrid search (BM25 + dense embeddings) improving contextual relevance and retrieval accuracy. • Applying LoRA/PEFT-based fine-tuning techniques optimizing LLM training efficiency and reducing compute cost by 60%. • Deploying scalable LLM inference services on Kubernetes enabling real-time inference with sub-150ms latency.
Citadel
May 20241 yr 3 mosJul 2025

Chicago, Illinois, United States

Developed predictive maintenance models using LSTM architectures for large-scale IoT telemetry datasets. • Built time-series forecasting pipelines improving anomaly detection precision by 30%. • Implemented transformer-based NLP models for log analysis and diagnostic intelligence. • Performed feature engineering and preprocessing for high-dimensional sensor datasets. • Applied hyperparameter tuning and cross-validation improving model accuracy and robustness. • Deployed ML models using AWS SageMaker with CI/CD automation workflows. • Containerized ML services using Docker and orchestrated deployments via Kubernetes. • Integrated MLflow for experiment tracking, model versioning, and reproducibility. • Monitored model performance and implemented automated retraining pipelines.
Aurobindo Pharma USA, Inc.
May 20212 yrs 3 mosJul 2023

Hyderabad, Telangana, India

• Designed scalable ETL pipelines processing multi-terabyte structured and semi-structured datasets. • Built distributed data processing workflows using Apache Spark for batch analytics. • Implemented data transformation pipelines enabling ML-ready datasets for downstream modeling. • Optimized SQL queries improving reporting performance by 40%. • Developed data warehousing solutions using PostgreSQL for analytics systems. • Implemented data validation frameworks ensuring 99.9% pipeline reliability. • Collaborated with analytics teams in preparing feature-engineered datasets for predictive models. • Automated data ingestion workflows improving processing efficiency. • Managed end-to-end data pipelines ensuring consistency across multiple data sources
Yashoda Hospitals
Aug 20182 yrs 9 mosApr 2021

Hyderabad, Telangana, India

• Developed backend systems and REST APIssupporting healthcare analytics applications. • Designed SQL-based reporting dashboards improving operational visibility. • Implemented data processing workflows for structured healthcare datasets. • Optimized database queries improving system performance and response time. • Automated reporting pipelines reducing manual effort by 30%. • Collaborated with cross-functional teams delivering scalable software solutions.

Education

Cleveland State University

Master's degree, Computer Science

Cleveland State University

20232 yrs2025

Join Sowmya and 10k+ other professionals on

Tal Boss

Tal is the app where bosses meet engineers & builders directly, without middlemen

Download Tal Boss

Sowmya Laveti on Tal

Headline

AI/ML Engineer

About

Senior AI/ML Engineer with 6+ years of experience specializing in production-grade Machine Learning Systems, Gen- erative AI, Agentic AI, and distributed ML infrastructure. Proven experience architecting end-to-end AI pipelines including Retrieval-Augmented Generation (RAG), LLM fine-tuning (LoRA/PEFT), agent-based workflows, real- time inference systems, and scalable cloud-native deployments. Strong expertise in MLOps, feature engineering, distributed data systems, model governance, and latency optimization. Demonstrated impact including 30% im- provement in anomaly detection precision, 35% reduction in inference latency, and 25% increase in predictive accu- racy across enterprise environments. Experienced in AWS, Kubernetes, Spark, PyTorch, TensorFlow, vector search systems, and multi-agent orchestration frameworks. Passionate about designing reliable, scalable, and cost-efficient AI platforms aligned with business strategy.

Location

Cleveland

Work experience

AI/ML Engineer at Lucid Motors, Aug 2025 - Present · 10 mos, Newark, California, United States, Architecting enterprise Generative AI systems integrating Retrieval-Augmented Generation (RAG) pipelines for internal knowledge and diagnostics platforms. • Designing embedding pipelines using transformer-based models and implementing FAISS vector indexing for se- mantic retrieval. • Implementing hybrid search (BM25 + dense embeddings) improving contextual relevance and retrieval accuracy. • Applying LoRA/PEFT-based fine-tuning techniques optimizing LLM training efficiency and reducing compute cost by 60%. • Deploying scalable LLM inference services on Kubernetes enabling real-time inference with sub-150ms latency.

Machine Learning Engineer at Citadel, May 2024 - Jul 2025 · 1 yr 3 mos, Chicago, Illinois, United States, Developed predictive maintenance models using LSTM architectures for large-scale IoT telemetry datasets. • Built time-series forecasting pipelines improving anomaly detection precision by 30%. • Implemented transformer-based NLP models for log analysis and diagnostic intelligence. • Performed feature engineering and preprocessing for high-dimensional sensor datasets. • Applied hyperparameter tuning and cross-validation improving model accuracy and robustness. • Deployed ML models using AWS SageMaker with CI/CD automation workflows. • Containerized ML services using Docker and orchestrated deployments via Kubernetes. • Integrated MLflow for experiment tracking, model versioning, and reproducibility. • Monitored model performance and implemented automated retraining pipelines.

Data Engineer at Aurobindo Pharma USA, Inc., May 2021 - Jul 2023 · 2 yrs 3 mos, Hyderabad, Telangana, India, • Designed scalable ETL pipelines processing multi-terabyte structured and semi-structured datasets. • Built distributed data processing workflows using Apache Spark for batch analytics. • Implemented data transformation pipelines enabling ML-ready datasets for downstream modeling. • Optimized SQL queries improving reporting performance by 40%. • Developed data warehousing solutions using PostgreSQL for analytics systems. • Implemented data validation frameworks ensuring 99.9% pipeline reliability. • Collaborated with analytics teams in preparing feature-engineered datasets for predictive models. • Automated data ingestion workflows improving processing efficiency. • Managed end-to-end data pipelines ensuring consistency across multiple data sources

Software Engineer at Yashoda Hospitals, Aug 2018 - Apr 2021 · 2 yrs 9 mos, Hyderabad, Telangana, India, • Developed backend systems and REST APIssupporting healthcare analytics applications. • Designed SQL-based reporting dashboards improving operational visibility. • Implemented data processing workflows for structured healthcare datasets. • Optimized database queries improving system performance and response time. • Automated reporting pipelines reducing manual effort by 30%. • Collaborated with cross-functional teams delivering scalable software solutions.

Education

Master's degree, Computer Science, Cleveland State University