Rajiv Vanipenta
Data Science, AI, ML
Hyderabad,TG
Summary
- A seasoned, technology-focused AI professional with 8+ years of progressive experience spanning the evolution from Data Science and Deep Learning to modern Generative AI, LLM engineering, and Agentic AI systems.
- Versatile contributor across the AI/ML lifecycle-from planning and architecture to development, deployment, and operations-building automated pipelines and production-grade systems that scale, drive operational efficiency, and deliver tangible business outcomes.
- Deep expertise in Generative AI technologies including LLMs, prompt engineering, few-shot learning, RAG (Retrieval-Augmented Generation) architectures, Agentic AI systems (Google ADK framework), MCP (Model Context Protocol), and Agent-to-Agent (A2A) communication protocols for creating intelligent, autonomous AI solutions.
- Specialized in developing predictive and analytical solutions using traditional ML models, ensemble methods, and deep learning architectures across diverse data types-structured, semi-structured, and unstructured-covering NLP, Computer Vision, Time Series forecasting, and cross-sectional analysis.
- Expert in deploying production-grade ML models with proven track record of building scalable REST APIs (FastAPI, Flask) for model serving, implementing CI/CD workflows, and establishing model monitoring systems for drift detection and performance tracking.
- Proficient in Google Cloud Platform (GCP) spanning AI/ML services (Vertex AI, Gemini/Gemini Enterprise, Document AI, Agent Builder, RAG Engine, Vertex AI Search, Dialog flow CX), data platforms (BigQuery, Bigtable, Cloud Storage, DLP), and deployment infrastructure (Cloud Run, GKE) for delivering scalable, production-ready AI applications.
- Proficient in Python-based ML ecosystem including 7+ years with Scikit-learn for classical ML, 6+ years with TensorFlow, 4+ years with PyTorch for deep learning, and comprehensive experience in hyperparameter tuning, model selection, and feature engineering.
- Skilled in Python, SQL, data engineering, DevOps, containerization (Docker, Kubernetes), and MLOps toolchains for end-to-end solution delivery, with strong foundation in system design principles and algorithmic problem-solving.
- Strong track record in managing client relationships, navigating stakeholder expectations, identifying and mitigating project risks, leading technical teams, mentoring interns and freshers, and applying strategic foresight to ensure timely, high-quality delivery.
- Cross-industry domain knowledge in Life Sciences, Healthcare, Fintech product development, E-commerce, and Smart Cities with IoT-enabled real-time governance, delivering tailored AI/ML applications that address sector-specific challenges and regulatory requirements.
Overview
10
10
years of professional experience
4
4
Certification
6
6
years of post-secondary education
Work History
Senior Technical Lead
Birlasoft Inc.
04.2022 - Current
- Senior Technical Lead for Johnson & Johnson's MedTech Pharmacovigilance operations, leading AI/ML and Generative AI initiatives that reduced adverse event processing time by 50% and automated 70% of ICSR case triage. Delivered intelligent solutions for Global Medical Safety (GMS), MedDRA coding automation, Product Quality Complaints, and ArisG case management, processing over 500K+ safety reports annually while ensuring FDA and EMA regulatory compliance.
- Platform Leadership: Spearheaded AI/ML solution development on Google Cloud Platform, specializing in Vertex AI services for Generative AI, Document AI, Healthcare API, and BigQuery to transform Pharmacovigilance operations.
- AI Innovation: Designed and deployed intelligent automation systems using Google Gemini (2.5, 2.0, 1.5), PaLM 2, and fine-tuned models for adverse event processing, ICSR data extraction, MedDRA coding, and case triage, significantly reducing manual effort.
- Solution Architecture: Built end-to-end pipelines for document processing, multilingual translation, medical text analysis, AI chatbots, and automated reporting, integrating Document AI, Healthcare API, Translation API, and custom NLP models.
- MLOps Implementation: Designed and deployed end-to-end MLOps infrastructure using MLflow for experiment tracking and model registry, Airflow for workflow orchestration, and Docker/Kubernetes for containerized model deployment, enabling reproducible and scalable ML operations across 15+ concurrent projects.
- Model Deployment Pipelines: Built automated CI/CD pipelines for ML model deployment using Vertex AI Pipelines and Kubeflow, implementing versioning, A/B testing, canary deployments, and automated rollback mechanisms for production models serving 1000+ predictions per hour.
- Production Monitoring & Observability: Established comprehensive monitoring systems using Vertex AI Model Monitoring, Cloud Logging, and custom dashboards to track model performance metrics, feature drift, prediction latency, and system health, reducing production incidents by 40%.
- Technical Ownership: Responsible for MLOps implementation, model governance, production monitoring, data security, and regulatory compliance across all AI deployments.
- Team Engagement & Mentorship: Supervised and mentored cross-functional teams of 15+ AI/ML Engineers, Data Engineers, conducting technical code reviews, knowledge-sharing sessions, and capability development in Generative AI, GCP architecture, and healthcare AI best practices.
- Strategic Planning & Governance: Developed project charters, technical roadmaps, feasibility studies, and capability assessments for AI initiatives, collaborating with clinical SMEs, product teams, and executive leadership to define requirements, timelines, and success metrics.
- Stakeholder Engagement: Served as technical advisor to C-level executives and business stakeholders, delivering architecture presentations, POC demonstrations, ROI analyses, and regulatory compliance assessments to drive strategic decision-making and secure project funding.
- Collaborated with clinical research scientists, medical and tech professionals specializing in Pregnancy & Lactation Safety, Pediatric Medicine, and Investigational Medicinal Products (IMP) to develop domain-specific AI solutions addressing critical patient safety and drug surveillance requirements.
- Key Projects & Implementations: EASE Framework - Regulatory Compliance Automation Platform, MLOps Platform & Model Deployment Infrastructure, Cognitive Document Processing - Intelligent Case Data Extraction, Methods and Analytics Hub - Conversational BI Chatbot, Regulatory Document Change Tracking & Analysis, Medical Contract Review & Classification, Late Case Prediction & Process Optimization, Signal Detection & Time Series Forecasting, MedDRA Auto-Coding for Adverse Events, Molecular Data Chatbot for Drug Research, Paediatric & Pregnancy Safety Data Hub, Machine Translation Pipeline for Safety Documents, Medical Literature Summarization, AIOps - Generative AI Operations Dashboard, Personalized Executive News Intelligence.
Hitachi Vantara
10.2021 - 03.2022
- Connected Systems for Urban Mobility: A Reinforcement learning approach using Deep Q Networks for optimization of scheduling of electric buses for better services to citizens and decreasing costs to the service operators.
Quantela Inc.
05.2020 - 10.2021
- Solid Waste Management & Vehicle Routing Optimization: Developed a mechanism by grouping the city into wards by predicting the amount of waste dumped at different time intervals and creating a platform that can optimally route the collection of vehicles thereby maintaining the society cleaner and reducing fuel consumption by vehicles. This system was proved to predict the garbage levels almost with 70% accuracy over a year. The process led to reducing the vehicle travel time by nearly 55% when compared over the preceding, thereby generating additional savings for the city administrators.
- Digitalization of Handwritten Forms - OCR: Manual entry of contact tracing data collected by the filed volunteers for the Covid patients is automated with a workflow to store the data in a central data storage portal. This was able to save 2000+ hrs. of manual work for digital entry of records, thus saving time and generating additional savings. The CER and WER achieved over handwritten forms in 22% and over computerized document are 7%.
- Safe City Preventive Policing: Call data of the Dial 100 is analyzed over a period for a district police authority to create a model that results in the propensity of a type of crime happening at a given place under a jurisdiction, with an intention to reduce the crime rate.
- Simulated Smart Traffic Management Systems: Simulation model of traffic density at junctions and important checkpoints in for a city at different time intervals at multiple possible conditions, constraints, looking at the historical and present situations.
- At Quantela, we work with various administrative bodies of cities, states situated across India and abroad with the vision of creating smart cities.
Random Trees LLC.
01.2018 - 04.2020
- Built app store for corporate clients with pre-defined AI / ML solutions for example document extraction, customer retention scores, fake invoice detections, text-to-speech, sentiment analysis, named entity extractions, task specific bots, Geo specific analytics which are deployed along with integration of various cloud ML service platforms such as AWS, GCP, Azure, IBM Cloud etc.
- This product is built on a Cortex AI platform, generating revenues based on a pay-per-use pricing model.
- Developed and deployed 25+ production ML models using Scikit-learn, TensorFlow, and PyTorch for document extraction, customer retention prediction, fraud detection, and sentiment analysis, serving 100+ enterprise clients via scalable REST APIs.
- Built automated ML pipelines using Airflow for data ingestion, feature engineering, model training, evaluation, and deployment across AWS, GCP, and Azure platforms, processing 10M+ records daily.
- Implemented MLflow for experiment tracking, model versioning, and registry management, enabling reproducible model development and seamless production deployment with full lineage tracking.
- Designed FastAPI-based microservices architecture for ML model serving with automatic scaling, load balancing, and health monitoring, achieving 99.9% uptime.
- AI Global Marketplace.
- Demand based pricing model for e-commerce travel platforms: Customer segmentation using data of more than 5 million customer data combined with demand across various routes, a dynamic pricing model by forecasting the demand for each customer & route segment thereby increasing profits for the operators. The up-graded marketing campaign launched by targeting a customer based on the customer group has seen a significant improvement in the click-through ratio of advertisements. With a segment label-based Interpretability of over 80% groups. Demand forecasts were generated with a RMSE score of ±3.
CSC (present DXC)
09.2013 - 03.2016
- 3R Health Insurance for Hospital Chain: Developed data pipelines of gathering data from multiple OLTP systems performing aggregations, transformations on the raw data and migrating it in a central OLAP data warehouse. Data quality and consistency is maintained over various data storage systems, which is used in preparation of reports and audits.
Education
BTech - Electronics and Communications
SRM University
09.2009 - 04.2013
Master of Science - Data Science, Machine Learning
Indian School of Business (ISB)
Hyderabad 10.2016 - 01.2018
Master of Science - Cloud Computing
Great Lakes
Executive 01.2021 - 03.2022
Skills
- LLM Engineering
- Gemini
- Claude
- GPT
- Gemma
- Agent Frameworks
- Google ADK
- LangChain
- LangGraph
- RAG
- Retrieval-Augmented Generation
- Prompt Engineering
- Few-shot Learning
- Multi-shot Learning
- Fine-tuning
- RLHF
- Vector Databases
- FAISS
- Vertex AI Vector Search
- Embeddings
- Text-embedding-ada-002
- Vertex AI Embeddings
- Sentence Transformers
- Protocols
- MCP
- Agent-to-Agent Communication
- AI Coding
- Gemini Code Assist
- Github-Copilot
- Gemini Antigravity
- ML/DL Frameworks
- TensorFlow
- PyTorch
- Keras
- Scikit-learn
- JAX
- Ensemble Methods
- XGBoost
- LightGBM
- CatBoost
- Random Forest
- Gradient Boosting
- Deep Learning
- CNN
- RNN
- LSTM
- Transformers
- Attention Models
- BERT
- Transfer Learning
- Computer Vision
- OpenCV
- PIL
- TensorFlow Object Detection
- YOLO
- Mask R-CNN
- Faster R-CNN
- NLP Libraries
- Hugging Face Transformers
- SpaCy
- NLTK
- Gensim
- Data Processing
- Pandas
- NumPy
- SciPy
- Dask
- Polars
- Visualization
- Matplotlib
- Seaborn
- Plotly
- Bokeh
- Streamlit
- Specialized
- Time Series Analysis
- Prophet
- ARIMA
- Anomaly Detection
- Recommender Systems
- Workflow Orchestration
- Apache Airflow
- Kubeflow Pipelines
- Prefect
- Vertex AI Pipelines
- Experiment Tracking
- MLflow
- TensorBoard
- Vertex AI Experiments
- Model Serving
- TensorFlow Serving
- TorchServe
- FastAPI
- Flask
- Containerization
- Docker
- Docker Compose
- Kubernetes
- CI/CD
- Bitbucket actions
- Jenkins
- JFrog
- Monitoring
- Logging
- Vertex AI Cloud Monitoring
- Cloud Logging
- Version Control
- Git
- GitHub
- DVC
- API Development
- API Gateway
- Rate limiting
- Swagger
- Model Monitoring
- Drift Detection
- Feature Distribution Tracking
- Performance Metrics
- Automated Retraining
- Cloud Platforms
- AWS Cloud Bed Rock
- SageMaker Studio
- GOOGLE CLOUD PLATFORM
- GCP
- AI/ML Services
- Vertex AI
- Model Training
- Deployment
- Pipelines
- Workbench
Websites
Certification
- Google Cloud Professional Machine Learning Engineer, 2023
- AWS Certified Machine Learning - Associate, 2019
- Pharmacovigilance Certified Professional - Life Sciences, 2023
- Anthropic Model Context Protocol Advanced Certification, 2025
- TensorFlow Developer Specialization - DeepLearning.ai, 2019
Personal Information
Total Experience: 10+ Yrs
Timeline
Senior Technical Lead
Birlasoft Inc.
04.2022 - Current
Hitachi Vantara
10.2021 - 03.2022
Master of Science - Cloud Computing
Great Lakes
01.2021 - 03.2022
Quantela Inc.
05.2020 - 10.2021
Random Trees LLC.
01.2018 - 04.2020
Master of Science - Data Science, Machine Learning
Indian School of Business (ISB)
10.2016 - 01.2018
CSC (present DXC)
09.2013 - 03.2016
BTech - Electronics and Communications
SRM University
09.2009 - 04.2013
Affiliations
- HydPy
Work Availability
monday
tuesday
wednesday
thursday
friday
saturday
sunday
morning
afternoon
evening
swipe to browse
Languages
English
Advanced (C1)