Pankaj Kumar Porwal

CEO

India19 yrs 4 mos experience
AI EnabledHighly Stable

Key Highlights

  • Led AI/ML strategy impacting multiple business areas.
  • Achieved significant cost savings through innovative solutions.
  • Expert in deploying scalable ML solutions in production.
Stackforce AI infers this person is a Data Science leader with expertise in AI/ML solutions across Fintech and E-commerce sectors.

Contact

Skills

Core Skills

Machine LearningData ScienceArtificial Intelligence (ai)

Other Skills

GenAICatBoostFraud DetectionTask AllocationOptimizationLead ScoringAuction OptimizationXGBoostRecommendation SystemsNLPStatistical ModelingMonte Carlo SimulationComputational ModelingApplied ProbabilityStatistics

About

• Leading the AI/ML strategy roadmap, identifying high-impact AI/ML and GenAI use cases across fraud, personalization, pricing, negotiation, customer support, and search & discovery, influencing product roadmaps and organizational investments in AI. • Collaborating with engineering and product teams to integrate ML models in customer-facing applications, improving user engagement and operational scalability. • Expertise in supervised/unsupervised learning, ensemble methods (XGBoost, LightGBM, CatBoost), recommendation systems, optimization techniques, reinforcement learning, deep learning (CNNs, Transformers) and NLP (GenAI, RAG, prompt engineering, embeddings, LangChain) • Proficient in Python (scikit-learn, pandas, numpy, PyTorch, Tensorflow), SQL (Snowflake, MySQL), Knowledge Graphs, Git, cloud-native workflows on GCP (Cloud Run, Cloud Function, GCS) and AWS (S3, Lambda) • Experienced in MLOps best practices with a track record of deploying and scaling ML solutions in production using cloud-native architectures (GCP, AWS), containerized services (Kubernetes, KServe), and robust CI/CD pipelines. Familiar with implementing observability, model monitoring, and orchestration using Airflow • Hands-on experience deploying ML models via REST APIs, batch (cron) jobs, and streaming pipelines (RabbitMQ, Pub/Sub); serving models on VMs with Gunicorn and process management with Supervisor • Strong foundation in experimental design (A/B testing) and causal inference, with hands-on experience applying these methods to optimize user experiences and drive business outcomes.

Experience

19 yrs 4 mos
Total Experience
3 yrs 8 mos
Average Tenure
8 mos
Current Experience

Amazon

Applied Science Manager - Alexa International

Oct 2025Present · 8 mos

Cars24

Director - Data Science (DS/ML Labs)

Oct 2021Oct 2025 · 4 yrs · Gurugram, Haryana, India · Hybrid

  • As a full-stack Data Science leader, I drive end-to-end execution of AI/ML solutions—from business problem identification and model development to production deployment and continuous performance monitoring
  • 🛡️ Fraud Detection & Dynamic Token System: Created a CatBoost-based fraud model and rule engine to predict dealer backout risk. Introduced dynamic token payment requirements as a risk-sharing strategy, boosting transactions while mitigating dealer backout risk resulting in ₹36 cr in annual savings.
  • 🔧 Task Allocation Engine: Designed and deployed an in-house engine to allocate vehicle inspection tasks to field experts, replacing a vendor system. This solution unified batch and on-demand flows with dynamic reallocation, leading to ₹2 cr in annual savings, 20% reduction in travel distance.
  • 📊 Lead Scoring & Funnel Optimization: Built CatBoost models to dynamically assess lead quality, enabling early filtering of low-intent users and smarter appointment slot prioritization. Improved conversion efficiency of call center and retail engagement efforts, driving a 12% uplift in Lead-to-Procurement (L2P) funnel performance.
  • ⚙️ Auction Ecosystem Optimization: Developed dynamic logic to tune auction levers—anchors, durations, and docking—for balancing auction coverage and competition. Resulted in a 15% increase in external bid participation and 4% uplift in top bid values.
  • 🎯 Dealer Recommendation Engine: Engineered a content-based recommendation engine using XGBoost to personalize auction listings based on dealers’ click, bid, and purchase patterns. Delivered a 4× lift in impression-to-view and impression-to-bid metrics for the top 100 cars.
  • 🎙️ AI-Powered Multilingual Support: Developed an AI-driven solution utilizing fine-tuned Whisper on Indian languages and GenAI (ChatGPT) to transcribe, translate, and summarize customer calls, extracting overall intent disposition. Improved customer support efficiency by delivering actionable insights, enhancing response accuracy.
Machine LearningData ScienceArtificial Intelligence (AI)GenAI

Techno india njr institute of technology, udaipur

Principal

Jul 2010Oct 2021 · 11 yrs 3 mos · On-site

  • 🎓 Educator | Innovator | AI/ML Practitioner
  • Experienced in teaching and mentoring across Artificial Intelligence, Machine Learning, Deep Learning, Applied Mathematics, and Mechanics. I have delivered hands-on workshops and academic courses while guiding hackathons, startups, and research projects—many of which led to patents and practical industrial impact.
  • 🔬 Innovation & IP
  • US Patent (2023): Co-inventor of a low-cost, ventilator using intelligent control systems—optimizing compression dynamics via real-time pressure feedback. Designed to be robust and usable in field/ambulance settings.
  • Indian Patent (2019): Developed a novel technique to manufacture bricks, paver blocks, and construction products using 100% waste materials—plastic, fly ash, marble slurry, and demolition debris—providing a sustainable alternative to conventional clay and concrete solutions.
  • 💡 Startup & Hackathon Involvement
  • Actively mentored startups and innovation challenges, bridging academic insights with scalable tech implementations. Passionate about applying research to solve real-world problems in healthtech, sustainability, and infrastructure.

Indian institute of technology, bombay

Assistant Professor

Jun 2007Jun 2009 · 2 yrs · Mumbai, India

Cornell university

Post Doctoral Associate

Dec 2006May 2007 · 5 mos

Los alamos national laboratory

Graduate Research Assistant

Aug 2004Aug 2005 · 1 yr

Education

Cornell University

Doctor of Philosophy (PhD)

Jan 2001Jan 2006

Indian Institute of Technology, Bombay

Bachelor of Technology (BTech)

Jan 1997Jan 2001

Indian Institute of Technology, Bombay

Stackforce found 100+ more professionals with Machine Learning & Data Science

Explore similar profiles based on matching skills and experience