Feb 2025 – Present · 1 yr 1 mo · Bengaluru, Karnataka, India · On-site

• Beyond 5G – NAS & Device Commercialization
• Working in the Beyond 5G team on the NAS (Non-Access Stratum) layer, supporting the commercialization of Samsung devices such as the M Series, A Series, S25, and Tab models for the North American market. Responsible for analyzing and resolving network-related issues across major operators including AT&T, Verizon, CHA, and CHT. Hands-on experience with the IQI module, field log analysis, and VOC (Voice of Customer) handling to ensure smooth carrier certification and user experience.
• Building internal automation tools to streamline workflows related to issue resolution, CPR (Change Package Request) identification, release tracking, and TR (Test Report) generation. Working across the full stack using React.js for the frontend and Flask for the backend, aimed at reducing manual effort and improving efficiency within the commercialization process.

Jul 2022 – Mar 2024 · 1 yr 8 mos · Bengaluru, Karnataka, India

• Image Translation – Samsung AI Feature
• Developed an advanced Image Translation feature as part of Samsung’s AI suite, designed to seamlessly translate text within images into the user’s target language. Performed in-place translation while preserving font color, size, alignment, and styling. Used inpainting techniques to remove original text and blend translated content into the background, delivering a native, overlay-free visual experience.
• ShotSuggestion – Camera Intelligence (Commercialized Across Samsung Devices)
• Contributed to the ShotSuggestion feature, rolled out across A Series, M Series, S Series, Tab, Fold, and Flip devices. Enabled intelligent framing guidance on the camera preview, helping users capture cleaner, more professional-looking photos. Integrated ML models:
• all_mnet.tflite for real-time object detection (~80ms)
• nima_crop.tflite for aesthetic quality scoring (~14ms)
• Implemented a pattern detection engine to disable suggestions in repetitive or noisy backgrounds, improving shot relevance.
• CI/CD & KPI Automation – ImageTranslation & ShotSuggestion Projects
• Led CI/CD pipeline integration on Jenkins, SonarQube, with tools like SAM, TEM, TCM, SVACE, and RA, all on a Linux-based codebase.
• Wrote JUnit tests for four branches (dev, devJar2.0, feature/hybridrendering, feature/multithread) reaching up to 80% coverage.
• Configured CTC++ code coverage for ShotSuggestion with all KPIs successfully passed.

Jan 2022 – Jul 2022 · 6 mos · Bengaluru, Karnataka, India

• Worked on the problem statement of image/photo deblurring after sharing multiple times on SMS.
• Proposed a synthetic degradation pipeline to reduce the effect of blur.
• Tune 9 hyper-parameters for training i.e isotropic blur, random kernels, etc. with probabilistic values in a range.
• Use DIV2K and Flickr2K datasets for training.
• Deeply understands the working and limitations of GANS (SRGAN, ESRGAN, Real-ESRGAN).
• Finally, proposed a light weighted highly optimized model to minimize blur for smart devices like Smart Phones.