Feb 2025 – Oct 2025 · 8 mos · Hyderabad, Telangana, India · On-site

• Designed and developed backend services for Azure Tenant Scanner (AzTs), a critical platform that safeguards Microsoft's vast digital estate by continuously scanning over 10M resources daily for cloud security risks — including misconfigurations, malware, and leaked secrets.
• Built micro service for graph-based enrichment layer to intelligently correlate and prioritize risks, reducing alert fatigue by surfacing the most critical security issues — important toxic combinations & attack paths, cutting down noisy alerts by 50%
• Led the design of a GenAI-powered remediation agent to provide accurate, contextual fixes with 100% precision and safety grounding as a key design goal — improving cloud security posture & MTTR by 23%
• Tech Stack: C#, .NET, Azure Cloud, Kusto (KQL), Python, CI/CD, Micro services, ETL Stream processing, Neo4J, Cypher

Aug 2020 – Aug 2024 · 4 yrs

• As CTO and co-founder of a D2C fashion startup, I applied GenAI, ML, Software Engineering and data-driven systems to scale the brand from scratch to eight figure valuation (3.2 Cr) in 3 years. Here's a sneak peak of some challenges we overcame -
• Customer queries piling up faster than we could answer — Built an LLM-powered chatbot that automated 90% of support tickets, improving response times and cutting manual load to near-zero.
• Inventory planning killing cash flow and shelf space — Developed ML-driven forecasting models that predicted demand with precision, reducing deadstock and enabling lean operations.
• Burning money on digital ads with little return — Applied intelligent clustering on Facebook pixel data to segment audiences and ran personalized ad campaigns, driving 1 Cr+ cost-efficient traffic and securing #1 Google rankings within 4 months.
• Too slow to drop fresh designs and chase trends — Used GenAI tools like Midjourney to 10× our design pipeline, launching high-quality collections at startup speed.
• Tech stack:
• Java, Spring boot, AWS cloud, Docker, Kubernetes, CI CD, Python, Tensorflow, Huggingface, Langchain, Langgraph, Crew AI, MLflow, Apache Airflow, Spark, Flink, ETL & data pipelines

Jun 2019 – Nov 2019 · 5 mos · Greater Hyderabad Area

• Built an ML-driven intelligence system for Microsoft Outlook – Systems processing over 10M emails in to tackle inbox clutter by classifying emails into meaningful categories (e.g., Flights, Hotels, Promotions) and extracting key details, enabling smarter email organization and user-focused insights.
• Designed large-scale, privacy-compliant data pipelines – using K-anonymization and PII-safe sampling techniques, incorporating novel hashing methods to securely process sensitive user email data.
• Trained supervised machine learning models – for both classification and information extraction, with a strong focus on robust feature engineering to boost model accuracy and generalization across diverse email formats, improving P1 scores by 21%.
• Deployed real-time ML inference services – with strict low-latency requirements, ensuring seamless integration into Outlook with responsive, high-confidence predictions.
• Tech Stack: C# .NET, Azure Cloud, Python, Tensorflow, Scikit learn, COSMOS, Hive

Jun 2016 – Jun 2019 · 3 yrs · Greater Hyderabad Area

• Development of pipelines for classification and smart organization of Outlook emails using Machine Learning & NLP algorithms.
• Android Developer in the team that built SMS Organizer app.

May 2015 – Jul 2015 · 2 mos · Bangalore

• Built an app for Android Wearable devices in Intuit Turbotax Product.