‹ Back
Senior ML Engineer - Big Entities
JOB SUMMARY
Roles
Job details
Our client Big Entities is looking for Senior ML Engineer to work remotely.
Location
: DHA Phase 3 (Remote)Timings: 10 AM - 7 PM (Fri - Sat Off days)Role SummaryWe are looking for a hands-on Computer Vision and ML Engineer with deep expertise in deep object detection and strong production delivery skills.
You will own end-to-end detection systems, from dataset and training pipelines to optimized inference services, monitoring, and continuous improvement. Key ResponsibilitiesDesign, train, debug, and improve state-of-the-art object detection models for real-world conditionsBuild robust training pipelines: datasets, augmentation, caching, versioning, and reproducible experimentsPerform systematic error analysis and ablations to isolate failure modes (data vs model vs inference vs post-processing)Develop custom detection systems beyond standard training, including multi-stage pipelines, ensembles, and specialized post-processingOptimize inference for latency, throughput, and memory, including GPU acceleration and export toolchainsDeliver production-grade services using Docker, Linux, CI/CD, and APIs (FastAPI and/or gRPC)Implement testing strategy across the pipeline (unit, integration, regression), including golden image test setsSet up monitoring and maintenance: logging, metrics dashboards, drift/performance tracking, retraining triggersWrite clear technical documentation, architecture decisions, and trade-off analysesRead research papers and rapidly translate ideas into working prototypes and deployable componentsRequired Skills and Experience:Python and ML EngineeringAdvanced Python engineering: clean architecture, packaging, typing, testing, profilingStrong PyTorch experience (must)TensorFlow optionalStrong model debugging skills and disciplined experimentationExperiment tracking and reproducibility: W B and/or MLflow, deterministic runs, seed controlConfig management: Hydra and/or OmegaConfData pipelines: PyTorch Dataset/DataLoader, augmentation pipelines, cachingDataset versioning: DVC or equivalentComputer Vision FundamentalsStrong CV fundamentals: preprocessing, geometry, photometric effects, distortions, camera modelsOpenCV expertise for classical CV and integration into modern ML pipelinesEvaluation expertise: mAP, precision/recall, IoU, PR curves, calibrationDeep Object Detection ExpertiseHands-on experience with modern detectors such as: YOLO (v5/v8/v9), Faster R-CNN, RetinaNet, EfficientDet, DETR variantsExperience building advanced detection workflows:Multi-stage detection (proposal, refine, classify)Ensemble and stacking strategiesSpecialized post-processing tuned to domain constraints
Product
ion ML and MLOps DeliveryModel export and serving: ONNX export/runtime, plus at least one of TorchScript or TensorRTGPU inference optimization and performance tuning (batching, throughput, latency, memory)Deployment: Docker, Linux, CI/CD basics (GitHub Actions and/or GitLab CI)Service implementation: FastAPI and/or gRPC, model versioning, rollback strategyMonitoring and lifecycle: drift/performance monitoring, logging, dashboards, retraining triggersTesting: unit tests for preprocessing/post-processing, integration tests, regression sets, threshold stability testsR D CapabilityAbility to read papers and implement ideas quicklyStrong debugging methodology, ablation design, and error analysisClear technical writing and engineering decision-makingNice-to-Have (Strong Bonuses)Engineering Drawings DomainExperience with engineering drawings and technical documentsPDF vector vs raster workflows, line detection, symbol detectionTable/diagram understanding, CAD-like concepts, annotation workflowsOCR + vision hybrid systems (even if not OCR-first)Document and Diagram Vision ToolchainPyMuPDF and/or pdfplumberImage rasterization, coordinate transformsHandling noisy scans: skew/warp correction, deskewingBroader CV CapabilitiesInstance segmentation: Mask R-CNN, YOLO-segKeypoints, pose, landmark detectionTracking for video: ByteTrack, DeepSORT.
Location
: DHA Phase 3 (Remote)Timings: 10 AM - 7 PM (Fri - Sat Off days)Role SummaryWe are looking for a hands-on Computer Vision and ML Engineer with deep expertise in deep object detection and strong production delivery skills.
You will own end-to-end detection systems, from dataset and training pipelines to optimized inference services, monitoring, and continuous improvement. Key ResponsibilitiesDesign, train, debug, and improve state-of-the-art object detection models for real-world conditionsBuild robust training pipelines: datasets, augmentation, caching, versioning, and reproducible experimentsPerform systematic error analysis and ablations to isolate failure modes (data vs model vs inference vs post-processing)Develop custom detection systems beyond standard training, including multi-stage pipelines, ensembles, and specialized post-processingOptimize inference for latency, throughput, and memory, including GPU acceleration and export toolchainsDeliver production-grade services using Docker, Linux, CI/CD, and APIs (FastAPI and/or gRPC)Implement testing strategy across the pipeline (unit, integration, regression), including golden image test setsSet up monitoring and maintenance: logging, metrics dashboards, drift/performance tracking, retraining triggersWrite clear technical documentation, architecture decisions, and trade-off analysesRead research papers and rapidly translate ideas into working prototypes and deployable componentsRequired Skills and Experience:Python and ML EngineeringAdvanced Python engineering: clean architecture, packaging, typing, testing, profilingStrong PyTorch experience (must)TensorFlow optionalStrong model debugging skills and disciplined experimentationExperiment tracking and reproducibility: W B and/or MLflow, deterministic runs, seed controlConfig management: Hydra and/or OmegaConfData pipelines: PyTorch Dataset/DataLoader, augmentation pipelines, cachingDataset versioning: DVC or equivalentComputer Vision FundamentalsStrong CV fundamentals: preprocessing, geometry, photometric effects, distortions, camera modelsOpenCV expertise for classical CV and integration into modern ML pipelinesEvaluation expertise: mAP, precision/recall, IoU, PR curves, calibrationDeep Object Detection ExpertiseHands-on experience with modern detectors such as: YOLO (v5/v8/v9), Faster R-CNN, RetinaNet, EfficientDet, DETR variantsExperience building advanced detection workflows:Multi-stage detection (proposal, refine, classify)Ensemble and stacking strategiesSpecialized post-processing tuned to domain constraints
Product
ion ML and MLOps DeliveryModel export and serving: ONNX export/runtime, plus at least one of TorchScript or TensorRTGPU inference optimization and performance tuning (batching, throughput, latency, memory)Deployment: Docker, Linux, CI/CD basics (GitHub Actions and/or GitLab CI)Service implementation: FastAPI and/or gRPC, model versioning, rollback strategyMonitoring and lifecycle: drift/performance monitoring, logging, dashboards, retraining triggersTesting: unit tests for preprocessing/post-processing, integration tests, regression sets, threshold stability testsR D CapabilityAbility to read papers and implement ideas quicklyStrong debugging methodology, ablation design, and error analysisClear technical writing and engineering decision-makingNice-to-Have (Strong Bonuses)Engineering Drawings DomainExperience with engineering drawings and technical documentsPDF vector vs raster workflows, line detection, symbol detectionTable/diagram understanding, CAD-like concepts, annotation workflowsOCR + vision hybrid systems (even if not OCR-first)Document and Diagram Vision ToolchainPyMuPDF and/or pdfplumberImage rasterization, coordinate transformsHandling noisy scans: skew/warp correction, deskewingBroader CV CapabilitiesInstance segmentation: Mask R-CNN, YOLO-segKeypoints, pose, landmark detectionTracking for video: ByteTrack, DeepSORT.
Discover the company
Explore other offers from this company or learn more about Taraki.
The company
T
Taraki Pakistan
