cv

Masters in Technology (M.Tech.)

Malaviya National Institute of Technology, Jaipur

• Specialization in Computer Science and Engineering (CGPA - 8.75)
• Awarded Gold medal for outstanding academic performance.
• Worked on Face Ageing using Style Transfer thesis under guidance of Dr. Neeta Nain, sponsored by MeitY, Gov. of India.

Bachelors in Technology (B.Tech.)

Dr. Babasaheb Ambedkar Technological University, Lonere

• Specialization in Computer Science and Engineering (CGPA - 8.9)
• Class Representative for 4 years
• Member of organizing committee for TEDxBATU, Cynosure Techfest.

Software Engineer — Embedded Systems Runtime and Firmware

Google (Silicon / Pixel Hardware), Bangalore, India

• Led Edge AI Runtime Silicon Bring-Up: Drove EdgeTPU (NPU) and Neural HAL enablement for next-generation Google Tensor mobile SoCs, resolving early-silicon dynamic linker faults, platform detection logic, and kernel security permissions to achieve stable on-device ML execution.
• Spearheaded Fleet-Wide NNAPI Deprecation: Authored technical justification and executed platform-level decommissioning of legacy NNAPI vendor HALs across Android; verified OS stability and quantified a 15.2x latency surge and 81% CPU thread stall across top 1P/3P applications to accelerate ecosystem transition to LiteRT.
• End-to-End Bug Triage Platform Implementation: Implemented an automated bug triage service completely from the ground up using Google Cloud Spanner distributed locking to eliminate race conditions across microservices; reduced manual issue duplication by 50%, automated crash artifact/ramdump processing, and saved 5–10 engineering hours/week.
• Zero-Overhead Kernel Hang Detection: Designed an atomic RAII tracker with 100% coverage over DSP/ISP kernel driver interactions integrated with watchdog diagnostics; resolved critical runtime issues on Pixel phones, improving overall system stability by 30%.
• Camera Test Automation & HW Verification: Built automated testing suites using Python and ADB covering all camera algorithms and use cases; developed low-level verification tests for critical hardware behaviors (DMA buffers, power management/DVFS), optimizing multi-firmware performance benchmarking.
• Telemetry Profiling & Memory Optimization: Synchronized userspace memory (showmap RSS/PSS), kernel DMA buffers (dmabuf_dump), and CPU cycles (simpleperf) for in-depth system trace analysis; developed C-API controls for dynamic memory pool sizing and automated TCM allocator resets.
• Dynamic Tensor Execution in LiteRT: Engineered support for dynamic tensor dimensions in LiteRT delegate kernels, enabling hardware-accelerated on-device execution for variable-input vision and language models.

Software Developer Associate

Halliburton, Bangalore

• Implemented Landmark Secrets Management (LSM) to automate the rotation of security credentials in AWS and Azure cloud environments
• Developed an error notification system using Python and NerdGraph GraphQL API of New Relic, deploying AWS Lambda and Azure Functions via Terraform.
• Developed an automated customer onboarding system by integrating Okta API with Python and Bash scripting, significantly reducing the workload on the support team by 80%
• Actively participated in regular Scrum meetings, collaborating with cross-functional teams to troubleshoot and resolve challenges in a dynamic software development environment.

Team Lead DS365.ai Support, Cloud Support Analyst II

Halliburton, Bangalore

• Led a dynamic support team for the "DS365.ai" platform, ensuring high-quality customer interactions and issue resolution
• Developed and implemented new support strategies, including the creation of comprehensive knowledge base articles, reducing recurring inquiries by 30%.
• Analysis of issues logged by end users, business team and working on incidents and severity issues
• Designed and developed custom reports tailored to user issue analysis utilizing Salesforce Lightning Experience's reporting capabilities.

Eklaavya Summer Internship

Indian Institute of Technology, Bombay (IITB)

• Developed Physics Interactive Animation Creator, an interactive platform to create physics animations with minimal coding.
• Responsible to create physics apparatus using Three.js library.

MuSTAT Face Ageing using Multi-Scale Target Age Style Transfer

CVIP, IIT Jammu

• This work proposes a multi-scale target age-based style face ageing model using an encoder-decoder architecture to generate high-fidelity face images under ageing
• Proposed using skip connections with selective transfer units (STU) to select and modify the encoder feature and used style information gathered from a random image of the target age group to train the generator.

Face Ageing using Style Transfer

• A project under Khoj Apno Ki | Tracing Missing Children, sponsored by Meity, Govt. of India.
• Proposed a method to synthesize images with age progression and rejuvenation effects based on given target style.
• Trained generative models using PyTorch and made comparative studies with baseline models.

Text Summarizer

• Developed a web app to generate a short summary of textual input using Extractive algorithms using Python packages Flask, Numpy and NLTK

Content Based Image Retrieval

• Created a web app using Python Flask framework to retieve images from database similar to image query (Query-by-image) based on color, texture and shape.
• Used OpenCV and Tensorflow to implement the model and MongoDB database to store image information.

• Gold Medal in M.Tech

• Best Paper Award for MuSTAT Face Ageing using Multi-Scale Target Age Style Transfer at CVIP, IIT Jammu.