Key ResponsibilitiesComputer Vision Pipeline DevelopmentDesign and implement real-time CV pipelines for object detection, tracking, and classification meeting Build multi-object tracking systems across camera feeds with re-identification and trajectory forecastingDevelop preprocessing pipelines for video streams (frame extraction, normalization, augmentation) with error handling and backpressure mechanismsImplement annotation workflows and active learning loops to continuously improve model qualityModel Engineering OptimizationFine-tune and evaluate SOTA open-source models (YOLO, EfficientDet, DETR families) on domain-specific datasetsOptimize inference throughput: batching strategies, model quantization (INT8/FP16), ONNX/TensorRT conversion, and multi-GPU orchestrationBuild A/B testing frameworks to measure model performance (mAP, FPS, recall@IOU) in productionMaintain model registry with versioning, lineage tracking, and rollback capabilitiesProduction ML InfrastructureArchitect scalable ML services exposing REST/gRPC APIs with authentication, rate limiting, and circuit breakersContainerize models and services (Docker) with CI/CD pipelines for automated testing and deploymentImplement monitoring dashboards tracking inference latency, GPU utilization, prediction confidence distributions, and data driftOwn incident response: debug production issues, conduct root-cause analysis, implement permanent fixesSoftware Engineering ExcellenceWrite maintainable Python code with type hints, unit/integration tests (pytest), and API documentationDesign clear data contracts between services; validate schemas with Pydantic/protobufConduct thorough code reviews focusing on performance, maintainability, and ML best practicesDocument system architecture, model cards, and operational runbooksCollaboration MentorshipPartner with data engineers on annotation tooling, dataset pipelines, and feature storesWork with DevOps to optimize Kubernetes deployments, autoscaling policies, and cost efficiencyMentor junior engineers on CV fundamentals, debugging techniques, and production ML patternsPresent technical deep-dives to cross-functional stakeholdersMinimum QualificationsEducation: Bachelors in Computer Science, Computer Engineering, Electrical Engineering, or related fieldExperience: 3-6 years building and deploying ML systems in production environmentsComputer Vision: Proven track record shipping CV solutions (object detection, segmentation, tracking, or pose estimation) handling real-world dataPython Proficiency: Strong software engineering skills—clean code, testing (pytest/unittest), packaging, virtual environments, type hintsModel Deployment: Experience serving models via REST/gRPC APIs with frameworks like FastAPI, Flask, or TorchServeInfrastructure: Hands-on with Docker, CI/CD tools (GitHub Actions, GitLab CI), and cloud platforms (AWS/Azure/GCP) or on-prem GPU clustersPerformance Tuning: Practical experience profiling code (cProfile, py-spy), optimizing memory usage, and reducing inference latencyPreferred QualificationsMasters degree in Computer Science, Data Science, Machine Learning, or related fieldAdvanced CV: Multi-object tracking (SORT, DeepSORT, ByteTrack), trajectory forecasting, or video understanding modelsModel Serving: Experience with Triton Inference Server, TorchServe, vLLM, or TensorRT optimizationsLLM/RAG Systems: Built retrieval-augmented generation pipelines using vector databases (Pinecone, Weaviate, Milvus) and embedding modelsEdge Deployment: Optimized models for edge devices (NVIDIA Jetson, Coral TPU) with latency/power constraintsMLOps Maturity: Worked with experiment tracking (MLflow, Weights Biases), feature stores (Feast, Tecton), or Kubernetes operators (KubeFlow, Seldon)Distributed Training: Experience with multi-GPU training (DDP, DeepSpeed) or large-scale data processing (Ray, Dask)