Advanced AI & Decision Intelligence
Designing production‑grade AI systems that power high‑stakes decisions in enterprise environments.
Designing production‑grade AI systems that power high‑stakes decisions in enterprise environments.
I design and lead the development of scalable, explainable, and governance‑ready AI systems. My work bridges machine learning, operations research, and modern AI architectures—including generative and agent‑based systems— with a strong emphasis on real‑world reliability.
Projects organized by broad data science categories: image, text, tabulated analytics, and time-series solutions.
Broad Category: Image + Text + Tabulated Analytics
Role: Senior Delivery Manager, AI Tower
Duration: Feb 2021 – Mar 2022
Project Details: Built a complete AI-enabled physician consultation platform for a health insurance provider, including telehealth scheduling, video conferencing, real-time alerts, and diagnostics.
Key Work: X-ray/CT disease detection, pathology and vitals analytics, live video anxiety detection, nail-image health screening, NLP chatbot for FAQs, and vital sensor alerting for post-operative discharge.
Technology: Python, TensorFlow, Flask, React, AWS Lambda, Glue, SageMaker, NLP, computer vision.
Learn More: View project page
Broad Category: Image
Role: Delivery Head / Principal Data Scientist
Duration: Jan 2018 – Jan 2021
Project Details: Developed advanced computer vision pipelines for satellite and mobile mapping imagery, and 3D LiDAR asset detection for transportation, utilities and infrastructure clients.
Key Work: Road marking and building footprint extraction, land-cover classification, POI detection, face/number-plate blurring, and LiDAR-based vegetation and wear detection using PointNet/Open3D.
Technology: Python, TensorFlow, Open3D, PointNet, Azure DataBricks.
Learn More: View project page
Broad Category: Tabulated Analytics + Time Series
Role: Principal Data Scientist
Duration: Jan 2018 – Jan 2021
Project Details: Delivered a predictive maintenance platform for 150K+ elevators and escalators across USA, Germany and Spain, using operational and historical maintenance data.
Key Work: Created daily elevator-level features, built supervised models for failure prediction, and delivered alerts and dashboards for maintenance teams.
Technology: Azure Databricks, Data Lake, ADF, Python, PySpark, Tableau, Power BI, Azure CI/CD.
Learn More: View project page
Broad Category: Time Series + Text
Role: Principal Data Scientist
Duration: Jan 2018 – Jul 2019
Project Details: Built prognostic engines for aeroengine combustion distress prediction and diagnosis, combining sensor time-series data with text extraction capabilities.
Key Work: Combustion distress forecasting, real-time diagnostics, and document/image-based text extraction for asset information capture.
Technology: Python, ML models, image text extraction, advanced analytics.
Learn More: View project page
Broad Category: Tabulated Analytics + Optimization
Role: Principal Data Scientist / Lead Data Scientist
Duration: 2015 – 2018
Project Details: Delivered optimization and analytics solutions across coal procurement, training scheduling, freight routing, and telecom asset management.
Key Work: Mixed-integer procurement planning for metal manufacturing, training management optimization for airline operators, demand forecasting for FMCG clients, crew scheduling, and telecom alarm management.
Technology: Python, GLPK, Java, optimization solvers.
Learn More: View project page
Academic research and master’s projects with a focus on optimization and decision science.
Broad Category: Tabulated Analytics + Optimization
Project Details: Developed stochastic models for transporting perishable items under uncertainty, using chance-constrained programming to balance cost, time, and decay.
Key Work: Designed algorithms for multi-objective stochastic transportation, implemented MATLAB prototypes, and validated performance for food supply-chain networks.
Learn More: View project page
Broad Category: Tabulated Analytics + Optimization
Project Details: Researched scheduling methods for parallel batch processors, minimizing total weighted tardiness and completion time with ant colony optimization.
Key Work: Created metaheuristics, compared results with existing benchmarks, and implemented solutions in C and LINGO for large-scale scheduling problems.
Learn More: View project page
Broad Category: Tabulated Analytics + Optimization
Project Details: Surveyed scheduling problems in grid computing and proposed metaheuristic architectures to optimize resource allocation in distributed systems.
Key Work: Developed solution architecture, evaluated scheduling algorithms, and presented findings during internship at GE John F. Welch Technology Centre.
Learn More: View project page
Demonstrations of technical capabilities and tools.
Description: Showcase of mathematical equation rendering using LaTeX and MathJax.
Examples: Knapsack problem optimization equations and linear regression formulations.
Technology: MathJax, LaTeX, HTML/CSS.
View Sample: LaTeX Equations Demo
Email: dr.ramana@outlook.com
Phone: +1 202 509 7562
Location: Glen Allen, Virginia