AI Powered Order Fulfillment
Scaled AI fulfillment support across operations, accelerating workflows by 60% and unlocking $5M+ in savings while eliminating 300+ daily support sessions.
AI Powered Order Fulfillment
Scaled AI fulfillment support across operations, accelerating workflows by 60% and unlocking $5M+ in savings while eliminating 300+ daily support sessions.
AI Powered Order Fulfillment
Scaled AI fulfillment support across operations, accelerating workflows by 60% and unlocking $5M+ in savings while eliminating 300+ daily support sessions.
Status & Year
Awarded (2024-25)
Status
Awarded (2024-25)
Status & Year
Awarded (2024-25)
Company
Walmart
Company
Walmart
Disclaimer
This project is based on real-world experience in a walmart's enterprise environment. Sensitive data, system identifiers have been modified to maintain confidentiality.
Problem Statement & Challenges
Operational teams at Walmart navigated 200+ fragmented modules to fulfill orders, relying heavily on manual decision-making.
This complexity slowed execution, increased errors, and created a strong dependency on training.
Annual training costs ranged from $774–$1,047 per associate, contributing to $5M+ in operational friction driven by cognitive overload, navigation inefficiencies, and communication delays.
My Contribution
I led the end-to-end design, research, and launch of an AI-powered fulfillment assistant that automated repetitive tasks and improved decision quality through transparent, explainable reasoning.
Through user sessions and workflow analysis, I defined the conversational strategy, interaction model, and safety guardrails partnering with Product and Engineering to prototype and operationalize autonomous fulfillment at scale.
Design
1 Member
Design
1 Member
Product
2 Members
Product
2 Members
Engg+Data+Ops
3 Members
Engg+Data+Ops
3 Members
Timeline
2 Sprints
Timeline
2 Sprints
Persona, Jobs to be done
Across 16+ sessions, a clear pattern emerged: Teams weren’t lacking effort they were overwhelmed by system fragmentation. They needed automation for routine tasks, faster decision cycles, lower training dependency, and human-like clarity in complex scenarios.
Fulfillment team:
Navigate 200+ operational modules daily.
Need intelligent, consolidated tools that reduce task overhead and onboarding dependency.
Ops team:
Handle escalations under time pressure.
Require instant, accurate information to resolve issues without workflow disruption.
Store managers:
Oversee end-to-end efficiency at scale.
Depend on real-time insights to optimize cost, throughput, and service quality.
Our North Star
Reduce operational expenditure while accelerating fulfillment through real-time, intelligent support that improves productivity, accuracy, and customer experience.
Field interaction
~16 Interviews
Research
~16 Interviews
Field interaction
~16 Interviews
System logs
~300 Sessions
System logs
~300 Sessions
Sample
5+ Stores
Sample
5+ Stores
Mapping & Analysis
5 days
Mapping & Analysis
5 days
Through natural turn-based interactions and built-in safety guardrails, it enabled faster, more accurate decisions at scale.
I designed a conversational AI system that unified order management across 200+ modules, replacing manual coordination with contextual, real-time guidance.
Ideate, Test & Measure
Designing AI systems requires clarity, transparency, and ethical guardrails especially when decisions affect cost, time, and customer trust. Explainability and human control are central to adoption.
Designing AI systems requires clarity, transparency, and ethical guardrails especially when decisions affect cost, time, and customer trust. Explainability and human control are central to adoption.
Future opportunities include role-based personalization, predictive recommendations, and applying the system to additional fulfillment domains.
From prototype to production operationalizing intelligent decision support at scale.
From prototype to production operationalizing intelligent decision support at scale.
The assistant was rolled out through phased deployment and live workflow validation across teams.
Workflows accelerated by 60%, 300+ daily support sessions were reduced, and the system unlocked $5M+ in operational savings.
What began as automation became a structural shift in how fulfillment decisions were made replacing fragmented coordination with unified, AI-assisted clarity.
Impact
$5 MN (Business savings)
Impact
$5 MN (Business savings)
Impact
- 300 (Sessions per day)
Impact
- 300 (Sessions per day)
Impact
4 Stars (Service rating)
Impact
4 Stars (Service rating)
Impact
<3 Mins (Task time)
Impact
<3 Mins (Task time)
Continue Reading...
AI Powered Order Fulfillment
Scaled AI fulfillment support across operations, accelerating workflows by 60% and unlocking $5M+ in savings while eliminating 300+ daily support sessions.
AI Powered Order Fulfillment
Scaled AI fulfillment support across operations, accelerating workflows by 60% and unlocking $5M+ in savings while eliminating 300+ daily support sessions.
AI Powered Order Fulfillment
Scaled AI fulfillment support across operations, accelerating workflows by 60% and unlocking $5M+ in savings while eliminating 300+ daily support sessions.
Status & Year
Awarded (2024-25)
Status
Awarded (2024-25)
Status & Year
Awarded (2024-25)
Company
Walmart
Company
Walmart
Disclaimer
This project is based on real-world experience in a walmart's enterprise environment. Sensitive data, system identifiers have been modified to maintain confidentiality.
Problem Statement & Challenges
Operational teams at Walmart navigated 200+ fragmented modules to fulfill orders, relying heavily on manual decision-making.
This complexity slowed execution, increased errors, and created a strong dependency on training.
Annual training costs ranged from $774–$1,047 per associate, contributing to $5M+ in operational friction driven by cognitive overload, navigation inefficiencies, and communication delays.
My Contribution
I led the end-to-end design, research, and launch of an AI-powered fulfillment assistant that automated repetitive tasks and improved decision quality through transparent, explainable reasoning.
Through user sessions and workflow analysis, I defined the conversational strategy, interaction model, and safety guardrails partnering with Product and Engineering to prototype and operationalize autonomous fulfillment at scale.
Design
1 Member
Design
1 Member
Product
2 Members
Product
2 Members
Engg+Data+Ops
3 Members
Engg+Data+Ops
3 Members
Timeline
2 Sprints
Timeline
2 Sprints
Persona, Jobs to be done
Across 16+ sessions, a clear pattern emerged: Teams weren’t lacking effort they were overwhelmed by system fragmentation. They needed automation for routine tasks, faster decision cycles, lower training dependency, and human-like clarity in complex scenarios.
Fulfillment team:
Navigate 200+ operational modules daily.
Need intelligent, consolidated tools that reduce task overhead and onboarding dependency.
Ops team:
Handle escalations under time pressure.
Require instant, accurate information to resolve issues without workflow disruption.
Store managers:
Oversee end-to-end efficiency at scale.
Depend on real-time insights to optimize cost, throughput, and service quality.
Our North Star
Reduce operational expenditure while accelerating fulfillment through real-time, intelligent support that improves productivity, accuracy, and customer experience.
Field interaction
~16 Interviews
Research
~16 Interviews
Field interaction
~16 Interviews
System logs
~300 Sessions
System logs
~300 Sessions
Sample
5+ Stores
Sample
5+ Stores
Mapping & Analysis
5 days
Mapping & Analysis
5 days
Through natural turn-based interactions and built-in safety guardrails, it enabled faster, more accurate decisions at scale.
I designed a conversational AI system that unified order management across 200+ modules, replacing manual coordination with contextual, real-time guidance.
Ideate, Test & Measure
Designing AI systems requires clarity, transparency, and ethical guardrails especially when decisions affect cost, time, and customer trust. Explainability and human control are central to adoption.
Designing AI systems requires clarity, transparency, and ethical guardrails especially when decisions affect cost, time, and customer trust. Explainability and human control are central to adoption.
Future opportunities include role-based personalization, predictive recommendations, and applying the system to additional fulfillment domains.
From prototype to production operationalizing intelligent decision support at scale.
From prototype to production operationalizing intelligent decision support at scale.
The assistant was rolled out through phased deployment and live workflow validation across teams.
Workflows accelerated by 60%, 300+ daily support sessions were reduced, and the system unlocked $5M+ in operational savings.
What began as automation became a structural shift in how fulfillment decisions were made replacing fragmented coordination with unified, AI-assisted clarity.
Impact
$5 MN (Business savings)
Impact
$5 MN (Business savings)
Impact
- 300 (Sessions per day)
Impact
- 300 (Sessions per day)
Impact
4 Stars (Service rating)
Impact
4 Stars (Service rating)
Impact
<3 Mins (Task time)
Impact