Viveka Kulharia

CEO

Seattle, Washington, United States4 yrs 9 mos experience

AI EnabledAI ML Practitioner

Key Highlights

8+ years in computer vision and generative AI.
Published at top-tier venues like CVPR and NeurIPS.
Proven track record in deploying state-of-the-art models.

Stackforce AI infers this person is a Computer Vision and Machine Learning expert with a focus on AI-driven automotive solutions.

Contact

vivekakulharia@gmail.com LinkedIn

Skills

Core Skills

Computer VisionMachine Learning

Other Skills

3D scene understanding3D shape predictionCC#C++Convolutional Neural Networks (CNN)Facial Expression RecognitionGANsGNU OctaveGPU processingHTMLJavaScriptLaTeXLinuxMicrosoft Office

About

Computer Vision researcher and engineer with 8+ years of experience in generative AI, diffusion models, and deep learning. I am interested in cutting edge research in computer vision and machine learning deployed for the real world applications. Published at top-tier venues (CVPR, ECCV, NeurIPS) with proven track record of deploying state-of-the-art models that improve real-world performance. Google scholar: https://scholar.google.co.in/citations?user=wjrOaIIAAAAJ

Experience

Moonvalley

Member of Technical Staff

Apr 2025 – Present · 11 mos · Seattle, Washington, United States

Research Engineer on the Data Intelligence team behind Marey, a state-of-the-art video generation model that surpassed OpenAI’s Sora 1 (T2V) at launch.
Ownership of end-to-end video data quality assessment, including metadata validation and correction by efficiently processing millions of videos using optimized GPU and CPU pipelines
Designed, implemented, and deployed state-of-the-art video understanding model
Lead the synthetic data post-processing pipeline and contributed to post-training dataset creation to improve model performance.

video generationdata quality assessmentGPU processingmodel deploymentComputer VisionMachine Learning

Cruise

Senior Applied Scientist

Oct 2022 – Apr 2025 · 2 yrs 6 mos · Seattle, Washington, United States · On-site

Generative AI (diffusion models, GANs), Perception, Machine Learning
Worked on perception systems for self-driving cars
Work spanned simulation data, generative AI, critical machine learning models for perception: https://www.infoq.com/news/2023/08/generative-ai-testing/
Improved perception model edge case handling (recall improved by > 30%) by building end-to-end
synthetic/real-to-real training data generation pipeline. Models deployed across two diﬀerent sensor
systems for self-driving perception.
Led the technical effort and cross-team alignment to integrate Generative AI into the perception stack
Proficiently utilized state-of-the-art generative methods like Diffusion models (https://arxiv.org/abs/2406.10722) and GANs (https://arxiv.org/abs/2303.12704) to
enhance perception system robustness
Had the unique opportunity to be mentored by Abhishek Sharma, Ashish Shrivastava, Bharat Singh, Luyu Yang, Arridhana Ciptadi, Avinash Ravichandran, Ambrish Tyagi, Huayan Wang, Taylor Lloyd, Gaspard van Koningsveld, Hang Zhang, and so many more inspiring individuals

generative AIdiffusion modelsGANsperception systemssynthetic data generationMachine Learning+1

Huawei

Senior Researcher, Computer Vision

May 2022 – Sep 2022 · 4 mos · Helsinki, Uusimaa, Finland · On-site

Worked with Dr. Ionut Cosmin Duta and Dr. Baiqiang Xia
Delivered production-ready neural network for similar image search on https://www.petalsearch.com/
Led data preparation, benchmarked diﬀerent neural networks to decide the candidate for our task,
created tools for very fast evaluations, coordinated with internal and external team members for project goal alignment, helped deliver the trained neural network for deployment

neural networksimage searchdata preparationComputer Vision

Niantic, inc.

Research And Development Intern

May 2021 – Aug 2021 · 3 mos · London, England, United Kingdom

Worked with Dr. Eric Brachmann, Dr. Aron Monszpart, Dr. Sara Vicente, Dr. Guillermo Garcia-Hernando, Prof. Gabriel Brostow
Developed novel neural network architecture for 3D scene understanding to target augmented reality
Created specialized dataset to support the research objectives

neural network architecture3D scene understandingdataset creationComputer Vision

Amazon lab126

Applied Scientist

Jun 2019 – Sep 2019 · 3 mos · Sunnyvale, California, United States

Worked with Dr. Siddhartha Chandra, Dr. Amit Agrawal, and Dr. Ambrish Tyagi
Developed state-of-the-art weakly supervised image segmentation method, accepted in ECCV 2020

weakly supervised learningimage segmentationComputer Vision

Technical university munich

Research Intern

Jun 2017 – Aug 2017 · 2 mos · Garching

Worked with Arnab Ghosh and Prof. Matthias Niessner
Created a model to predict coarse 3D shape and colors of an object given its single-viewpoint image
Rendered images and created voxel data from available meshes using MLib
Worked on getting finer and diverse predictions

3D shape predictionimage renderingvoxel data creationComputer Vision

Microsoft

Research Fellow

Jun 2016 – May 2017 · 11 mos · Bengaluru Area, India

Advised by Dr. Sundararajan Sellamanickam
Worked on Cloud Services Modeling. Designed anomaly detection models for multi-variate time-series
Explored MRF based approach to compare time-series of different types based on unusual characteristics
Created recommender system for Office application

anomaly detectiontime-series analysisrecommender systemsMachine Learning

Xerox research centre india

Research Intern

May 2015 – Jul 2015 · 2 mos · Bengaluru Area, India

Advised by Dr. Narayanan Unny
Explored lasso regression to get interpretable Sparse model for a high feature dataset
Created a novel method to estimate missing values under constraints

lasso regressionsparse modeling

Monet networks inc.

Software Developer

May 2014 – Jul 2014 · 2 mos

Advised by Dr. Anurag Bist, CEO
Understood the existing Facial Expression Recognition API and its usage. Worked on backend to capture and store video using existing WebRTC APIs
Developed specific metrics on non-verbal cue analytics for content rating

Facial Expression RecognitionWebRTC APIs