# Shiva Lakumarapu

> AI Engineer | Generative AI & LLM Systems | Prompt Engineering | NLP | LangChain | Hugging Face | AWS SageMaker | PyTorch | GPT-4

Location: Irving, Texas, United States
Profile: https://flows.cv/shivalakumarapu

I’m an AI/ML Engineer specializing in Generative AI, LLM systems, and intelligent automation.

Over the past 5+ years, I’ve designed and deployed large-scale machine learning and NLP solutions across finance, healthcare, and insurance — building everything from RAG-based knowledge assistants to transformer-powered compliance systems.

At Gilead Sciences, I develop LLM applications using GPT-4, BERT, and LangChain, integrating them with FAISS and pgvector for semantic retrieval and contextual grounding. My focus is on building production-ready GenAI pipelines with strong MLOps foundations using AWS SageMaker, MLflow, and Docker/Kubernetes.

I’m passionate about prompt engineering, multimodal AI, and explainable ML, ensuring every model I ship is interpretable, scalable, and business-ready.

⚙️ Core Skills: GPT-4 | LangChain | Prompt Engineering | RAG | Hugging Face | PyTorch | TensorFlow | AWS SageMaker | FastAPI | MLflow | NLP | Computer Vision | Docker | Kubernetes

📚 Recent focus: LLM-based document summarization, multimodal report generation, and generative AI observability.

## Work Experience
### Gen AI Developer @ Gilead Sciences
Jan 2024 – Present | Dallas, Texas, United States
Designing and deploying Generative AI systems using GPT-4, BERT, and LangChain for summarization, content generation, and conversational intelligence.
Implementing Retrieval-Augmented Generation (RAG) with FAISS, Pinecone, and pgvector for high-precision semantic retrieval.
Deploying large models on AWS SageMaker and Google Vertex AI, integrating continuous evaluation and drift monitoring via MLflow.
Building modular, containerized FastAPI microservices for AI inference using Docker and Kubernetes

### Data Scientist / AI-ML Engineer @ TD
Jan 2023 – Jan 2024 | New York, United States
Developed NLP pipelines for sentiment analysis and compliance report summarization using GPT-3, BERT, and T5.
Built semantic and vector-based search systems with FAISS and Pinecone for financial document intelligence.
Automated MLOps workflows using Airflow, Docker, and SageMaker for scalable retraining and CI/CD deployment.
Applied SHAP and LIME for explainable AI and integrated Grafana dashboards for production monitoring.

### Data Scientist / ML Engineer @ Allianz Partners
Jan 2020 – Jan 2022 | Hyderabad, Telangana, India
Developed scalable machine learning pipelines in Python and PySpark for churn prediction and risk analytics.
Built NLP systems using spaCy, Word2Vec, and TF-IDF for sentiment and intent classification.
Deployed APIs with Flask and Docker, orchestrated using Kubernetes for distributed inference.
Introduced MLflow-based model tracking and experiment versioning to streamline MLOps workflows.


## Education
### Master's degree in Computer and Information Sciences, General
University of North Texas


## Contact & Social
- LinkedIn: https://linkedin.com/in/sivamanikanta-l

---
Source: https://flows.cv/shivalakumarapu
JSON Resume: https://flows.cv/shivalakumarapu/resume.json
Last updated: 2026-04-18