Profile photo

Anubhav Dhawan

AI/ML Product Leader & Consultant

HomePortfolioResume

Let's Talk!

Commercial AI/ML products led from concept to launch.

Products

Predictive Pricing Engine

Product Lead & Builder
01
1
Goal

Stabilize and scale pricing operations by augmenting current pricing processes with an ML-powered pricing engine, restoring speed and consistency without requiring a replacement hire after the departure of the company's sole pricing expert.

2
Toolkit
TF-IDF
Logistic Regression
XGBoost
Decision Trees + Bagging
Scikit-learn
Vibe Coding
3
Bottom Line
90% reduction in manual effort for pricing and eliminated the need to setup a new pricing team in a cash strapped start up.
Accomplishments
  • Reduced reliance on manual workflows, improving efficiency and analyst productivity (~90% effort reduction)
  • Restored pricing velocity and eliminated quote backlogs following disruption in pricing operations
  • Maintained continuity without additional headcount, empowering business analysts to run pricing workflows, avoiding the need for a specialized pricing hire
  • Improved pricing consistency and margin reliability through standardized, model-driven decisioning
View Artifacts(undisclosed due to proprietary reasons.)
Products

On-Demand Reporting & Analytics (NL2SQL)

Product Lead
02
1
Goal

Design and deploy an AI-powered NL2SQL solution to handle high volumes of custom reporting requests across 45+ B2B customers, enabling fully user-driven report generation with minimal to no human intervention.

2
Toolkit
Agentic AI
NL2SQL
RAG
Advanced Prompt Engineering
LLM Harness / Guardrails
LLM as Judge
Observability framework
User Feedback loop
Reusable report layer
End-to-end security framework
3
Bottom Line
Drastically reduced the time for custom requests, from weeks to minutes, and removed more than half of the engineering effort needed for these reports.
Accomplishments
  • Reduced software engineering workload by ~60% by automating custom report generation through natural language queries
  • Shortened turnaround time for reporting requests from 2–4 weeks to ~15 minutes
  • Enabled end users to define report requirements directly, eliminating most back-and-forth and manual SQL translation
  • Introduced reusable saved reports, allowing users to build personalized, continuously updated dashboards
  • Scaled analytics support across 45+ customers without proportional increase in engineering effort
View Artifacts(undisclosed due to proprietary reasons.)
Products

Conversational Knowledge Platform

Product Lead
03
1
Goal

Design and deploy a conversational AI platform to serve two critical functions: enable B2B customers to get instant answers to operational questions, and accelerate onboarding and training of new associates through guided, context-aware assistance.

2
Toolkit
Agentic AI
RAG
Vector Database
Contextual Conversation
Hallucination Prevention
LLM as Judge
Observability Framework
Feedback Loops
3
Bottom Line
Saved around half a million yearly in customer support and training costs.
Accomplishments
  • Customer-facing (B2B support)
    • Reduced customer support load through self-serve query resolution
    • Improved response time and consistency across interactions
  • Internal (training & onboarding)
    • Accelerated onboarding of new associates with real-time guidance
    • Reduced dependency on trainers and documentation
  • Platform-level
    • Created a unified knowledge layer powering multiple workflows, saving business $0.5M annually
    • Scaled support, training without a proportional increase in engineering effort
View Artifacts(undisclosed due to proprietary reasons.)
Products

Human-AI Collaboration Transformation

Transformation Lead
04
1
Goal

Lead an organization-wide transformation to adopt AI-native ways of working by repositioning AI from a perceived threat to an enabler, focusing on augmenting human capability, accelerating delivery, and empowering teams to build faster and smarter without reducing workforce value.

2
Toolkit
Operating Model Redesign
Change Management
Stakeholder Alignment
Narrative and Perception
AI-assisted Development
Structured Prompt Engineering Playbooks
3
Bottom Line
Transformed the engineering operational model for the company to an AI-first operating model reducing delivery time to half.
Accomplishments
  • Organizational Adoption & Culture Shift
    • Successfully convinced a skeptical organization to adopt AI, even when initial sentiment viewed it as a threat
    • Reframed AI as a tool for acceleration, not replacement
  • Delivery Acceleration
    • Improved the speed of execution by 50% by embedding AI into day-to-day workflows
    • Enabled rapid prototyping and iteration, fully driven by business teams
  • Sustainable Operating Model
    • Established a repeatable, organization-wide framework for adopting AI responsibly and effectively
    • Built long-term trust in AI systems through guardrails, transparency, and continuous feedback mechanisms