Kushagra Wadhwa

Software Engineer

Bengaluru, Karnataka, India2 yrs 10 mos experience

AI EnabledAI ML Practitioner

Key Highlights

Achieved 80% reduction in AI model latency.
Designed high-performance backend systems at Goldman Sachs.
Collaborative approach in solving complex networking challenges.

Stackforce AI infers this person is a Backend-focused Software Engineer in Telecommunications and Fintech.

Contact

wadhwakushagra01@gmail.com LinkedIn

Skills

Core Skills

MicroservicesJavaHigh Performance Computing (hpc)Ai Model AccelerationAi Model OptimizationPythonWeb Development

Other Skills

6GARM ArchitectureAndroid DevelopmentArtificial Intelligence (AI)C#C++CMakeCUDAComputer NetworkingComputer VisionConvolutional Neural Networks (CNN)Core OSData AnalyticsData Plane Development Kit (DPDK)Data Structures

About

I am a software engineer with a strong focus on building scalable systems and optimizing performance across distributed platforms. At Goldman Sachs, I am contributing to the design and development of resilient, high-performance backend systems that powered critical financial workflows - emphasizing reliability, throughput, and low-latency execution. Previously at Samsung, I leveraged my expertise in distributed computing and high-performance systems to advance telecommunications technology. My work focused on optimizing AI model latency and enabling over-the-air (OTA) data simulations using NVIDIA’s Aerial Omniverse Digital Twin (AODT). In a recent project, I achieved an 80% reduction in AI model latency through model acceleration and quantization. I graduated from Delhi Technological University (DTU) in 2023 with a B.Tech, bringing strong problem-solving skills and a solid foundation in software engineering. My collaborative approach and technical acumen have consistently empowered teams to solve complex challenges in next-generation networking and large-scale system design.

Experience

2 yrs 10 mos

Total Experience

1 yr 11 mos

Average Tenure

11 mos

Current Experience

Goldman sachs

Software Engineer

Jul 2025 – Present · 11 mos · Bengaluru, Karnataka, India · On-site

Designed and implemented a critical feature for a high-traffic platform (10M+ daily requests), using Java (Spring), React, and MongoDB; enhanced core client workflows and improved overall system efficiency.
Developed and deployed backend services using microservice architecture, ensuring scalability, fault tolerance, and seamless integration across distributed systems.

JavaSpringReactMongoDBMicroservices

Samsung r&d institute india

3 roles

Senior Software Engineer

Mar 2025 – Jul 2025 · 4 mos · Bengaluru, Karnataka, India

Working on High Performance Distributed Computing systems
Optimised AI model latency by 10% using model acceleration and Quantisation.
Simulated OTA data transmission setup using NVIDIA’s Aerial Omniverse Digital Twin (AODT) on GPUs

High Performance Computing (HPC)AI model accelerationQuantizationNVIDIA Aerial Omniverse Digital Twin

Software Engineer

Aug 2023 – Mar 2025 · 1 yr 7 mos · Bengaluru, Karnataka, India

Team: Advanced RAN SW(6G Labs)/ Beyond 5G
Using NVIDIA Aerial CUDA accelerated RAN framework to execute the complete OTA calls from UE to Core Network where L1 is accelerated to execute on NVIDIA GH100 GPUs.
Optimization of AI model inference time executing on NVIDIA GPU using TensorRT library which brings latency reduction, optimization and quantization to the model along with the execution of model in optimized CUDA graphs
Authored a research paper for the Samsung Best Paper Award titled ‘Power Optimization in Fast packet processing system’
Exploration of native AI solutions to bring inference optimization for execution on x86 cores using Intel AMX and AVX. It includes optimization and quantization of CNN AI models using tools like openVino to int8 and bf16 data types and benchmark
the inference timing and vectorization efficiency.
Worked on rearchitecting the network protocol stack legacy code. Understanding the L2 Layer code, specifically Carrier
Aggregation, and proposed a new architecture for these modules.
Designed and implemented optimisation techniques on multi-threaded legacy code to improve network performance and efficiency.
Worked with industry experts in L2 layer realization and optimization for 6G THz test beds.
Gained strong understanding of RAN and L2 layer concepts, protocols, such as job-scheduling algorithms used in MAC
Leading a team of 5 members towards developing an unreleased cutting-edge technology for Samsung Flagship products.

NVIDIA Aerial CUDATensorRTAI model optimizationMulti-threaded developmentHigh Performance Computing (HPC)

SDE Intern

May 2022 – Jul 2022 · 2 mos

Worked on UI for rendering the client-side application to test the new 6G network being developed by Samsung.
Worked on developing a 6400x6400 HD application for all OS platforms using python frameworks like PyQt5.
Researched on the new 6G network, and its implementation possibilities.
Improved the current user-interface of the application deployed in testing, and made it cross-platform compatible.
Analysed and processed the high-speed raw data received over internet and developed an application to plot graphs in response to the speed of the data using libraries like Matplotlib and Numpy

PythonPyQt5MatplotlibNumpy

Kroop ai

Software Development Intern

Dec 2021 – Apr 2022 · 4 mos

Worked on React and Django framework to develop state-of-the-art company website
Wrote and integrated API’s in Django and used them in React on the frontend to make the website functioning.

ReactDjangoWeb Development