Senior computer vision researcher

Following selection criteria will be followed -5-15 YOEWorked on at least 1 Computer vision problem statement in the past 5 yearsHave built, trained and deployed models on production beforeComfortable with WFO in Bangalore or NoidaHappy to read research papers day in day out and train on terabytes of image data on big H100 clusterspreferably tier-1 college (IISc, IITs, NITs, BITS and others)About UsSite Recon is a B2 B Saa S platform transforming property measurements for landscapers and snow removal contractors in the $176 B US market. Using advanced AI on high-resolution aerial imagery, we’ve reduced measurement time from 1 week to 1 day, aiming for 1 hour with upcoming AI advancements.Role OverviewWe seek a skilled engineer to evaluate, enhance and/or transform our computer vision infrastructure from traditional CNN architectures to cutting-edge transformer-based models. This role demands expertise in model design, training, and optimization.You’ll have access to:A vast aerial imagery database (7 cm GSD) of 500,000+ properties across the U. S. since 2021.A team of 60 annotators mapping thousands of acres daily.A ready market of 350+ customers for immediate deployment.Powerful compute resources for rapid model training.This is a frontier challenge in computer vision and GIS. While solutions like Meta’s SAM offer basic raster segmentation, they lack the precision for creating high-fidelity, topologically consistent vectors essential for practical GIS applications. SAM struggles with occlusions (e.g., shadows, tree canopies) and produces "blobs" rather than architect-quality outputs.Our approach focuses on solving a constrained problem: using the world’s highest-resolution aerial imagery (7 cm) over U. S. urban areas with logical, repeatable patterns. By tackling this focused challenge, we aim to develop scalable templates to generalize automated extraction for broader GIS applications, similar to Waymo’s strategy in self-driving technology.Key ResponsibilitiesDesign and implement transformer-based architecture for semantic segmentation of aerial imageryDevelop efficient image-to-token and token-to-image conversion pipelinesCreate and maintain training datasets, including data cleaning, augmentation, and validationOptimize model training processes for distributed computing environmentsImplement efficient inference pipelines for production deploymentCollaborate with engineering team to integrate new models into existing infrastructureRequired Technical SkillsStrong foundation in computer vision and deep learning fundamentalsExtensive experience training transformer models from scratchExpert-level proficiency in Py TorchExperience with ONNX or Tensor RT model optimization and deploymentDeep understanding of distributed computing and parallel processingAdvanced Python knowledge, including multi-threading and multi-processing optimizationExperience with semantic segmentation tasksProven track record of handling large-scale data processingRequired Experience5+ years of hands-on deep learning experienceTrack record of successfully deploying computer vision models in productionExperience with vision transformer architecturesExperience optimizing models for production using ONNX/Tensor RTBackground in handling high-resolution satellite/aerial imagery preferredMasters/Ph D in Computer Science, Machine Learning, or related field preferredDesired QualitiesStrong mathematical foundation in deep learning conceptsExperience with model architecture design and optimizationProven ability to conduct independent research and stay current with latest developmentsExcellence in technical documentation and communicationSelf-motivated with a passion for solving complex technical challengesWhat Sets You ApartExperience with vision transformers specifically for segmentation tasksPublished research or contributions to open-source computer vision projectsExperience with high-performance computing environmentsBackground in geospatial data processingHands-on experience with model quantization and optimization using ONNX/Tensor RTExperience deploying optimized models in production environmentsWhy This Role MattersEvery day without improved segmentation costs us real business opportunities. We need someone who moves fast, thinks systematically, and delivers production-ready improvements quickly.