Jan 2024 – Dec 2024 · 11 mos · Mumbai, Maharashtra, India · On-site

• At Endovia, a sister company of Motilal Oswal (one of India’s largest brokerages), I worked as the first engineering hire to design and scale the firm’s algorithmic trading infrastructure from scratch. Over the course of a year, I transitioned the organization from single-strategy prototypes to a robust, multi-strategy trading platform managing millions in assets.
• Algorithm Development & Optimization: Designed and deployed 8+ proprietary trading strategies (options straddles, strangles, and trend-based), collaborating closely with traders. Optimized the backtest engine to reduce runtime from 6 hours to ~1 minute on a year’s worth of minute-level data, accelerating research cycles and strategy iteration.
• Order Management System (OMS): Architected a centralized in-house OMS capable of routing trades across multiple brokers, handling 25+ client portfolios with ₹210M+ in funds, and achieving sub-10ms latency per order. Integrated fault tolerance, redundancy, and failover APIs for market data resilience.
• System Scalability: Modularized the trading infrastructure into smaller services, enabling faster development of new strategies and smoother onboarding of engineers. The team scaled from a single developer (myself) to a group of 11 interns and 5 full-time engineers.
• Impact: Platform currently serves 25+ clients, delivering consistent 27%+ annual returns on deployed strategies, while maintaining reliability in high-volume market conditions.

Nov 2023 – Jul 2024 · 8 mos · Palo Alto, California, United States · Remote

• At Induced.ai, I worked on building scalable cloud infrastructure for an AI-first company backed by Sam Altman, focused on deploying distributed proxy networks to support advanced AI workflows. My role combined systems engineering, networking, and reliability engineering to ensure performance at scale.
• Distributed Proxy Infrastructure: Designed and optimized load balancing across 100+ proxy servers, ensuring smooth routing of web sessions generated by AI-driven backend systems. Improved reliability by reducing downtime by 70% and doubled system throughput.
• Custom Proxy Solutions: Transitioned from Raspberry Pi–based proxies to a more customized infrastructure, increasing network redundancy by 60% and lowering deployment costs by 30% per node.
• Scalable Session Management: Built custom session-handling logic to manage thousands of concurrent web clients with unique routing requirements, ensuring both stability and performance.
• Cost Efficiency & Performance: Hosted routing infrastructure locally to reduce dependency on third-party APIs, cutting costs and improving request-handling latency.
• Startup Environment: Operated in a fast-paced, high-stakes startup culture, collaborating directly with founders and senior engineers to iterate rapidly and deploy production systems supporting core AI services.

Jul 2023 – Feb 2024 · 7 mos · Mumbai, Maharashtra, India · On-site

• At NCAIR, I worked on applied AI/ML research projects at the intersection of aerospace, manufacturing, and IoT systems, collaborating with industry partners and researchers to solve large-scale real-world problems.
• Defect Classification at Scale: Developed a machine learning model for V-Guard Industries, using 20TB of multivariate time-series data from 12+ industrial sensors collected over 2 years. Designed pipelines for cleaning, feature extraction, and classification to identify manufacturing defects in real time.
• IoT Systems & Prototyping: Built IoT solutions leveraging ESP32 microcontrollers, integrating custom PCB design, 3D printing, and sensor interfacing for real-time data monitoring in industrial environments.
• Research & Teaching Support: Assisted Professor Asim Tewari in his industry-focused AI/ML course, providing technical mentorship and supporting applied learning projects for industry participants.
• LLM-based Tools: Contributed to internal research initiatives, building LLM-driven web scraping tools to automate data collection pipelines for material science and aerospace datasets.

Jul 2023 – Dec 2023 · 5 mos · Navi Mumbai, Maharashtra, India · Hybrid

• At Henkel, I worked on enterprise-scale digital transformation initiatives, building mobile applications to streamline manufacturing workflows and modernize employee training.
• Production Line Management App: Developed a Flutter-based app to digitize batching, testing, job sequencing, and handover processes in adhesive manufacturing lines. Deployed pan-India, centralizing data for analytics and reducing manual effort by 60% across production facilities.
• Digital Training & Certification Platform: Built an app for employee upskilling and quality assurance, used by 1000+ employees nationwide. Integrated role-based access for trainers and trainees, digital evaluations, and automated certification from levels 1–3. Reduced evaluation time by 70%, while promoting continuous employee learning.
• Enterprise Integration: Collaborated with Henkel’s IT and operations teams to ensure seamless deployment, enabling real-time analytics dashboards for plant managers and corporate leadership.

Dec 2022 – Jun 2023 · 6 mos · Mumbai, Maharashtra, India · Hybrid

• At Easocare, I developed backend systems for pharmaceutical warehouse management software, with a focus on optimizing medicine delivery logistics across pharmacies in Mumbai.
• Routing API Development: Built a custom routing API to optimize delivery schedules, vehicle utilization, and driver assignments. Implemented clustering algorithms to group delivery areas and incorporated driver–vehicle preferences for balanced workloads.
• Operational Impact: Reduced fuel costs by 25% and improved delivery efficiency by 20%, while promoting healthier driver work hours through intelligent scheduling.
• Infrastructure & Cost Efficiency: Hosted OSRM (Open Source Routing Machine) locally, eliminating dependency on costly third-party APIs and cutting routing costs to 1/60th of previous methods, while also improving response times for route generation.
• Scalable System Design: Integrated the routing system into the warehouse management platform, ensuring it could handle high-volume daily orders for large pharmaceutical distributors.

Dec 2021 – May 2022 · 5 mos · Mumbai, Maharashtra, India · Hybrid

• At Capgemini, I worked on applying AI to large-scale industrial optimization problems, building reinforcement learning solutions for enterprise clients in the FMCG sector.
• Reinforcement Learning for Job Shop Scheduling: Designed and developed a deep reinforcement learning (RL) environment to optimize job shop scheduling, enabling dynamic sequencing of production tasks across multiple machines.
• Predictive Production Planning: Incorporated future demand forecasts and machine maintenance schedules into the RL model, ensuring both efficiency and minimal downtime in production lines.
• Impact on Client Operations: Improved production efficiency by 15%, reducing bottlenecks and enhancing throughput for FMCG manufacturing plants.
• Enterprise Collaboration: Partnered with Capgemini’s consulting and client teams to align technical solutions with real-world production requirements, demonstrating the ability to translate advanced AI concepts into business value.

Apr 2020 – Sep 2020 · 5 mos · Mumbai, Maharashtra, India

• Worked with a team of skilled and taleneted proffesionals. Gained a lot of insight in this new domain.