Sep 2021 – Present · 4 yrs 6 mos · Abu Dhabi Emirate, United Arab Emirates

• Designed and deployed reinforcement learning–based product recommendation systems, optimizing personalization and engagement across multilingual user bases, including Arabic-speaking segments in fintech and e-commerce.
• Led product strategy initiatives integrating generative UX frameworks and dynamically generated content using LLMs and visual intelligence, tailored for localized experiences including Arabic-first interfaces.
• Built and fine-tuned domain-specific LLMs and VLLMs for conversational AI in finance, security, and marketing. Deployed multilingual chatbots with optimized performance for Arabic, improving client support and engagement in MENA-region deployments.
• Engineered Web3-integrated NLP pipelines by training LLMs on smart contract corpora with Arabic text preprocessing, enabling decentralized audit analytics and cross-border compliance checks.
• Delivered customer churn prediction models in high-volume environments, incorporating behavior and language-specific patterns for better retention across diverse user demographics, including Arabic-speaking cohorts.
• Developed NLP-based automation for accounting and bookkeeping workflows, incorporating Arabic OCR and tokenization modules to support bilingual financial document processing.
• Fine-tuned audio-language models (ASR + TTS) for Arabic and English speech processing. Deployed them in onboarding, call center, and KYC pipelines, significantly enhancing speech accuracy and latency in BFSI sector voice operations.

Sep 2019 – Sep 2021 · 2 yrs

• Sep 2019 – Sep 2021 | Singapore
• Developed and deployed CTR prediction algorithms for Grab Ads, leveraging user behavior features, content embeddings, and time-based decay models to improve click-through rates and ad targeting precision across Southeast Asia.
• Collaborated with cross-functional teams (marketing, product, data science) to integrate real-time ML predictions into advertising campaigns, improving monetization outcomes.
• Designed and implemented scalable data architectures for Grab Financial Group (GFG), enabling reliable ingestion, transformation, and serving of financial transaction data for underwriting, scoring, and offer personalization.
• Contributed to architecture modernization for ML pipelines including feature store design, model versioning, and A/B testing support for rapid experimentation.
• Defined key success metrics (CTR uplift, latency, retention impact) and participated in business reviews to align ML output with marketing and financial growth KPIs.
• Facilitated sprint planning and cross-team delivery milestones as part of Grab’s centralized ML platform team.
• Technologies: Python, Go, AWS, Azure, Apache Spark, Airflow, SageMaker, ClickHouse
• Domains: Ads ML, Finance ML, Personalization, User Growth, Real-time ML Ops

Dec 2017 – Sep 2019 · 1 yr 9 mos · Bengaluru Area, India

• Worked within Amazon's Search & Personalization team, contributing to core algorithms powering product discovery and relevance ranking across high-traffic e-commerce surfaces.
• Developed and optimized feature pipelines for search ranking models, incorporating real-time behavioral signals, query context, and catalog metadata.
• Co-implemented scalable offline training and online inference architecture for collaborative filtering and hybrid recommendation algorithms.
• Participated in delivery of A/B experiments for search re-ranking and personalized product carousels, collaborating with scientists, product managers, and UX researchers.
• Reduced model inference latency by optimizing endpoint throughput using native AWS infrastructure (SageMaker, Lambda, DynamoDB, CloudWatch).
• Wrote internal tooling to automate deployment checks, monitor feature drift, and maintain model health post-deployment.
• Contributed to Amazon’s internal ML experimentation platform for training/testing pipelines with configurable hyperparameters and rollout gating.
• Liaised with product and retail business teams to define KPIs such as precision@K, click-through rate (CTR), and add-to-cart lift, and incorporated them into model validation suites.
• Mentored two junior engineers and participated in hiring loop for Applied Scientists and SDE-1/2 roles.
• Technologies: Python, Scala, AWS SageMaker, DynamoDB, CloudWatch, EMR, S3, JIRA
• Focus Areas: Personalized Ranking, Feature Store Design, Model Monitoring, Cross-functional Delivery

Aug 2016 – Nov 2017 · 1 yr 3 mos · Bengaluru Area, India

Oct 2015 – Jul 2016 · 9 mos · Bengaluru Area, India

Apr 2015 – Oct 2015 · 6 mos · Pune Area, India

Mar 2013 – Mar 2015 · 2 yrs · Pune Area, India

Aug 2012 – Mar 2013 · 7 mos · Pune/Pimpri-Chinchwad Area

• Developed in-house Python-Django based REST Service for image data visualisation using computer vision techniques that is mainly focussed on mobile consumers as a part of end course project in CDAC.
• Develop User Interface for the Android Clients using native Android Library.
• Develop User Interface for the iOS Clients using native iOS Library.

Jun 2010 – Jul 2012 · 2 yrs 1 mo