About Me
The science of better decisions, at scale
What I Work On
Every day, tens of millions of packages move through one of the world's most complex logistics networks. Behind every delivery promise — every estimated arrival time, every routing decision, every capacity allocation — there's a science problem waiting to be solved.
That's where I work. As a Senior Research Scientist at Amazon, I design the optimization and forecasting systems that power last-mile delivery at global scale. My systems manage over a million concurrent forecasting models, process millions of packages daily, and operate across six regions worldwide. I think of it as operations research (~50%), machine learning (~30%), and systems architecture (~20%) — all in service of one goal: making the right decision, at the right time, at enormous scale.
My Path
My journey into operations research started with a question that has driven my entire career: how do you make the best possible decision when the system is too large and too uncertain for intuition alone?
I pursued that question through a PhD in Supply Chain and Operations Management at the University of Texas at Austin, where I studied under Professor Anant Balakrishnan. My doctoral research on subscription pricing for delivery services was published in Production and Operations Management — and the insights proved remarkably relevant to the industry I'd soon join.
Before Amazon, I spent three years as Lead Operations Research Scientist at BNSF Railway, where I optimized train routing across a 32,500-mile network spanning 28 states. Our work contributed to BNSF receiving the 2018 INFORMS Prize — the highest organizational honor in operations research. That experience taught me something textbooks don't: that the best science means nothing if it doesn't survive contact with the real world.
Beyond the Algorithms
When I'm not solving optimization problems at work, I'm usually thinking about them anyway — whether that's in the context of a good book, a conversation with a fellow researcher, or figuring out the most efficient route through a grocery store (occupational hazard). I'm originally from India, hold degrees from NIT Rourkela and UT Austin, and currently live in Kirkland, Washington with my family. If you ever want to talk about the beauty of integer programming or debate whether forecasting is art or science — I'm your person.
Areas of Research
Problem spaces I work in — from theory to deployed systems
Subscription Pricing for Free Delivery Services
Balakrishnan, A., Sundaresan, S., & Mohapatra, C.
Production and Operations Management, 33(4), 943-961, 2024
Key finding: Universal free-delivery subscriptions generate 33.7% more profit vs. paid delivery.
Read Paper →Enhanced Batch Computing Architecture for Consensus Planning in Large-Scale Supply Chains
US 12,499,399 B1
Supply Chain Optimization · 2025
View on Google Patents →Designing Storage Pods with Layers of Bins or Slots
US 12,504,281 B1
Warehouse Operations · 2025
View on Google Patents →Delivery Network Optimization
Designing algorithms that decide how millions of packages move through large-scale delivery networks — balancing density, cost, speed, and carrier capacity in real time.
- • Combinatorial optimization for package-to-route assignment at national scale
- • Geospatial cost estimation with orders-of-magnitude granularity improvements
- • Real-time decision systems processing millions of packages daily
Demand Forecasting & Planning
Building forecasting platforms that manage millions of concurrent prediction models across global regions, combining statistical, deep learning, and ensemble methods to drive operational planning.
- • Multi-horizon forecasting architecture spanning 6 global regions
- • Hierarchical reconciliation across station, region, and network levels
- • Short-horizon models for real-time capacity alignment during demand surges
Warehouse Operations
Applying optimization and machine learning to warehouse floor operations — from task sequencing and makespan minimization to physical storage design.
- • ML-driven task duration prediction combined with integer programming for scheduling
- • Combinatorial optimization for configurable storage unit design
- • Won Top 2 OR Paper Award at Amazon Machine Learning Conference 2022
Subscription & Pricing
Game-theoretic modeling of subscription plan design under retail competition — determining when universal free delivery outperforms tiered or paid alternatives.
- • Universal free-delivery subscriptions generate 33.7% more profit vs. paid delivery
- • Published in Production and Operations Management (2024)
- • Analytical framework for competing retailers with heterogeneous consumers
Railroad Network Optimization
Optimizing train routing, scheduling, and infrastructure health across BNSF Railway's 32,500-mile national network — solving capacity-aware problems at previously intractable scale.
- • Capacity-aware routing and scheduling across 28 states and 3 Canadian provinces
- • Large-scale models (10M+ variables, 20M+ constraints) with sub-second runtime
- • Sensor-based defect detection processing terabytes of real-time data
Systems & Applied Impact
Research deployed into production at scale
Package Selection Systems
Research Scientist / Lead Research ScientistA family of optimization systems that determine which packages to assign to which delivery stations and routes, maximizing network density while balancing cost, speed, and capacity constraints across the entire fulfillment-to-doorstep pipeline.
Press: Supply Chain Dive, Route Advisors | INFORMS 2022, INFORMS 2024 (Invited Panel)
Demand Forecasting Systems
Lead Research ScientistA large-scale forecasting platform managing concurrent prediction models across multiple global regions, combining classical statistical, deep learning, and tree-based ensemble methods with hierarchical reconciliation to drive capacity and staffing decisions.
Press: INFORMS 2024, POMS 2024 (Invited Tutorial)
Capacity Planning & Supply Chain Coordination
Lead Research ScientistCapacity Planning Under Uncertainty & Supply Chain Coordination
Systems that allocate delivery capacity under demand uncertainty and coordinate across supply chain stakeholders — from warehouse labor planning to network-wide resource balancing — ensuring service targets are met even as conditions shift.
Press: POMS 2024 (Invited Tutorial), INFORMS 2023
Railroad Network Optimization
Lead Operations Research ScientistRailroad Network Optimization — BNSF Railway
Optimized train routing and scheduling across BNSF Railway's 32,500-mile rail network spanning 28 states and 3 Canadian provinces. Solved capacity-aware routing at a scale previously considered intractable.
Press: INFORMS Prize 2018
Sensor Health Detection & Failure Analysis
Lead Operations Research ScientistSensor Health Detection & Failure Analysis — BNSF Railway
Built predictive models for detecting rail infrastructure sensor degradation and failure patterns across BNSF Railway's network, enabling proactive maintenance and reducing unplanned service disruptions.
Press: INFORMS 2018 (INFORMS Prize Year)
Awards & Recognition
INFORMS Prize
Awarded to BNSF Railway for pioneering integration of OR. Past recipients include Intel, UPS, IBM, and Disney.
Role: Lead OR Scientist on the optimization team
INFORMS Invited Panelist
Selected to represent Amazon Last Mile on a panel on Network Analytics at the premier operations research conference.
Role: Invited Panelist
Top 2 OR Paper, Amazon ML Conference
Warehouse operations optimization combining machine learning with integer programming.
Role: Research Scientist
Best PhD Research Award
IROM Research Symposium, University of Texas at Austin. Recognized for doctoral research on subscription pricing.
Role: PhD Researcher
PhD Scholarship
Supply Chain Management Center of Excellence, McCombs School of Business, UT Austin.
Role: PhD Student
Bonham Fellowship
McCombs School of Business, University of Texas at Austin.
Role: PhD Student
Dean's Fellowship
McCombs School of Business, University of Texas at Austin.
Role: PhD Student
Patent Portfolio
Granted
Enhanced Batch Computing Architecture for Consensus Planning in Large-Scale Supply Chains
GrantedUS 12,499,399 B1 · 2025
View on Google Patents →Designing Storage Pods with Layers of Bins or Slots
GrantedUS 12,504,281 B1 · 2025
View on Google Patents →Pending
Route Optimization (2 patents)
PendingFiled: 2021, 2023
Warehouse Operations (2 patents)
PendingFiled: 2022
Demand Forecasting (1 patent)
PendingFiled: 2024
Supply Chain Optimization (1 patent)
PendingFiled: 2023
Cost Estimation (1 patent)
PendingFiled: 2022
Multi-agent Coordination (1 patent)
PendingFiled: 2026
Service to the Field
Peer Review
Ad-hoc reviewer for Networks, ITOR, Production & Operations Management, and European Journal of Operations Research.
Competition Judging
Judge, INFORMS RAS Problem Solving Competition (2018, 2019). Reviewed solutions from 50+ international teams.
Mentorship
Technical mentor to 14 research and data scientists across multiple Amazon teams and global regions.
Timeline of Public Work
Talks, publications, patents, awards, and service
New patent filed — Multi-agent AI Systems (April 2026)
Reviewer, ITOR — flagship journal of IFORS (50+ member countries)
Reviewer, ACL TrustNLP 2026 — reviewed 3 submissions for premier NLP conference
CSS 2026 — Oral (Operations Research & Optimization) + Poster (Machine Learning)
<9% oral acceptance rate
Session Chair (Invited), INFORMS 2026 — AI and Decision Making track
Patent Granted — Enhanced Batch Computing for Consensus Planning (US 12,499,399)
Patent Granted — Designing Storage Pods with Layers of Bins (US 12,504,281)
Session Chair, POMS Annual Conference — Data Science, Analytics, and AI
Atlanta, GA
Presenter, POMS — Amazon Last Mile Logistics: Jurisdiction, Demand, Capacity, and Labor Planning
Atlanta, GA
Co-author, INFORMS — Multi-period Newsvendor for Capacity Planning
Atlanta, GA
AMLC — 2 Oral Presentations (Capacity Planning Workshop, Gen-AI Autonomous Agents Workshop)
<10% acceptance rate
CSS — Oral Presentation + Poster Presentation
<10% oral acceptance rate
Gen AI Science Fair — Top 4 AI Agent Applications
Gen AI Symposium — Poster Presentation
Published — "Subscription Pricing for Free Delivery Services" in Production and Operations Management
Vol. 33(4), 943-961
Session Chair, POMS Annual Conference — Data Science and Analytics
Minneapolis, MN
Invited Panelist, INFORMS — Network Analytics at Amazon Last Mile Logistics
Seattle, WA
Invited Tutorial Speaker, POMS — Optimizing Demand and Capacity Planning
Minneapolis, MN
Presenter, INFORMS — Error-Based Forecast Refinement
Seattle, WA
Co-author, INFORMS — 3 presentations (Logistics Operations, Multi-Period Newsvendor, Deliveries and Pickups)
Seattle, WA
AMLC — 2 Poster Presentations
Press coverage — "Free delivery plans can profit both retailers and customers"
Phys.org / UT Austin
Session Chair & Presenter, INFORMS — Amazon Last Mile Warehouse Operations
Phoenix, AZ
Co-author, INFORMS — 3 presentations (Mobile Fueling, Consensus Planning, Overlapping Jurisdictions)
Phoenix, AZ
AMLC — Poster Presentation
Patent Filed — Overlapping Jurisdictions and Route Optimization
Patent Filed — Scalable Consensus Planning for Large-Scale Supply Chains
Top 2 OR Paper Award — Amazon Machine Learning Conference (Oral)
<10% acceptance rate
AMLC — Poster Presentation
CSS — Oral Presentation
<10% acceptance rate
Presenter, INFORMS — Capacity and Coincidence Aware Package Assignment
Indianapolis, IN
Co-author, INFORMS — Newsvendor Approach Capacity Planning
Indianapolis, IN
Press coverage — "Amazon rolls out delivery route algorithm to reduce miles driven"
Supply Chain Dive
4 patents filed — delivery cost estimation, warehouse makespan, storage pods, oversized package storage
Presenter, INFORMS — Demand Forecasting for New Nodes in a Delivery Network
Anaheim, CA
Presenter, MSOM Conference (Peer-Reviewed) — Delivery Pricing Strategies
Singapore
Judge, INFORMS RAS Problem Solving Competition
50+ international teams
INFORMS Prize — BNSF Railway recognized for pioneering OR integration
Highest organizational honor in operations research
Presenter, INFORMS — Optimization Models for Block Re-design
Phoenix, AZ
Main Presenter, INFORMS RAS — BNSF OR Tools Showcase (Industry Demo)
Phoenix, AZ
Judge, INFORMS RAS Problem Solving Competition
Peer Reviewer — Networks: An International Journal
Presenter, INFORMS — Pricing Strategies for Competing Retailers
Houston, TX
Invited Speaker, Indian School of Business — Subscription Pricing Research
Hyderabad, India
Co-author, Invited Talk, IIM Bangalore — Membership Has Its Benefits
Bangalore, India
Presenter, MSOM Conference (Peer-Reviewed) — Subscription Pricing
Auckland, New Zealand
Presenter, INFORMS Annual Meeting — Subscription Pricing
Nashville, TN
Co-author, Invited Talk, Rutgers University — Subscription Pricing
Camden, NJ
Session Chair, INFORMS Revenue Management and Pricing Conference — New Frontiers of Revenue Management
Columbia University, New York, NY
Presenter, INFORMS RMP — Subscription Pricing
Columbia University, New York, NY
Presenter, INFORMS Annual Meeting — Subscription Pricing
Philadelphia, PA
Presenter, INFORMS Annual Meeting — Optimal Movement and Transshipment of Rail Freight Shipments
Philadelphia, PA
Presenter, POMS Annual Conference — Optimal Route Planning and Scheduling of Long-haul Freight
Washington, DC
Co-author, Invited Talk, Wharton School — Predicting Life Changing Events from Customer Interaction Data
Philadelphia, PA
Co-author, INFORMS Annual Meeting — Joint Pricing and Free-Shipping Decisions
San Francisco, CA
Presenter, INFORMS Annual Meeting — Shipment Dispatching on Capacitated Networks
Charlotte, NC
Interested in Collaborating?
Whether it's research, speaking, or just a good conversation about OR and AI — I'd love to connect.
Let's Talk