Oct 2021 – Oct 2024 · 3 yrs · San Jose, California, United States · Hybrid

• ● Developed network and security configuration management and control planes, including REST APIs for software- defined networking (SDN) in a hybrid cloud distributed software system.
• ●Engineered advanced networking features for hybrid cloud solutions (on-premises and public cloud platforms such as AWS and Azure) which included developing transit virtual private clouds and supporting Kubernetes pods on the same networking stack, micro-segmentation and security for containerized workloads (Cilium as CNI for Kubernetes workloads)
• ● Implemented Role-Based Access Control (RBAC) features to significantly enhance security, compliance, and operational efficiency by ensuring appropriate user access levels based on their roles.
• ● Automated unit testing processes and streamlined development workflows to improve software quality and reliability. This automation reduced manual testing efforts and increased overall productivity.
• ● Demonstrated proficiency in microservices-based distributed system architectural development, leveraging cutting-edge technologies to enhance system scalability, resilience, and performance.
• ● Led and collaborated with multiple cross-functional teams to ensure successful end-to-end delivery of projects, fostering effective communication and alignment across diverse stakeholders.

Oct 2019 – Oct 2021 · 2 yrs · San Jose, California

• Design and development of infrastructure around various protocols: OSPF, LDP FRR, SR, TILFA etc. in a docker container based distributed environment
• Implemented transactional and model driven APIs for install, update, manipulate, delete of the user configuration data and retrieve the operational data using netconf, C, Python, XML, yang data modelling, Redis database.
• Implemented and enhanced the script automation for unit testing of the networking systems.
• Implemented micro-loop avoidance feature for protected paths to achieve better switch-overs.
• Debugged and analysed several production issues ranging from traffic drops, high switchover cases, memory
• corruptions, functional issues in protocols using gdb, wireshark and ixia
• Scale optimization of code to handle plethora of routes in the ip over mpls ring network topologies.
• CPU profiling of code and daemons using gprof on scaled setup scenarios, fixing memory leaks and corrupitons captured via valgrind.

Aug 2018 – Apr 2019 · 8 mos · Chandigarh Area, India

• Developed a solution (ResDNN )for low-level computer vision image restoration task to denoise natural and X-ray images. Implemented the model in Kears and tensorflow, used MATLAB for image pre-processing.
• Outperformed state-of-the-art methods by 5% increase in similarity index for X-ray images by preserving image details by supervised learning of image/noise features using deep learning (CNN). It utilizes skip connections, residual learning and optimized values of parameters for efficient model training.
• Teaching Assistant in Data Mining using Python practicum.
• Publications:
• Deep Convolution Neural Network-based Denoiser for Mammographic Images.
• Springers’ Communication in Computer and Information Sciences | ISBN 978-981-13-9939-8 | Acceptance Rate: 18%
• Deep Residual Learning-based denoiser for Medical X-ray Images.
• IEEE Transactions on Image Processing | Under Review (TIP-20772-2019)
• ResDNN: Deep Residual Learning for Natural Image Denoising.
• IET Image Processing |Under Review (IPR-2019-0623)

Sep 2016 – Dec 2016 · 3 mos · Gurgaon, India

• Optimized sales-force compensation strategy of pharma client by data-driven insights, visualization and dashboard development in Tableau. Used regression, k-means clustering, etc. for optimizing several KPIs.
• Built end to end data pipeline - from data preparation to deployment.
• Coordinated between non-technical client and ZS team for requirement and feedback gathering.

Jan 2016 – May 2016 · 4 mos · Chandigarh Area, India

• Partnered with engineering, QAs, and developer teams while carrying out networking simulation study for optimizing IEEE 802.15.4 / Zigbee in OPNET in W
• Implemented amends to the algorithm for enhanced throughput with a 3% reduction in end-end delay and 36% increase in no. of successfully acknowledged packets in C++.
• Developed an internal portal using HTML/CSS, Javascript, MySQL for automating attendance process and work hour logging to enable employees to manage profiles, plan schedules, etc.
• Prepared & presented technical reports documenting SDLC