Job Title: AI/ML Engineer
AI/ML Engineer
Experience- 5-8 yrs
Location- Pune
Work Mode- Work from Office 5 Days a Week | Koregaon Park, Pune
ROLE OVERVIEW
We are looking for an AI/ML Engineer with 5-8 years of experience, including 3+ years in Computer Vision and Deep Learning, with hands‑on experience in building, optimizing and deploying production AI/ML solutions across Cloud and Edge environments.
KEY RESPONSIBILITIES
- Develop, train, fine‑tune and evaluate Computer Vision and Deep Learning models.
- Prepare and process image/video datasets, including annotation, augmentation and class balancing.
- Build solutions for Object Detection, Classification, Segmentation, Tracking, OCR, Pose Estimation and Action Recognition.
- Develop real‑time and batch inference and image/video processing pipelines.
- Improve model accuracy, latency and throughput through evaluation and optimization.
- Deploy and optimize models across Cloud, GPU and Edge environments.
- Develop AI/ML services and REST APIs and integrate models with backend platforms.
- Implement model monitoring, testing, logging, failure handling and rollback.
- Collaborate with Backend, Cloud, DevOps and Data Engineering teams.
- Troubleshoot production issues and maintain model and dataset documentation.
AI/ML & COMPUTER VISION
- Machine Learning, Deep Learning, Transfer Learning, Fine‑Tuning and Hyperparameter Tuning
- Model Evaluation, Precision, Recall, F1 Score, Confusion Matrix and Error Analysis
- Image Classification, Object Detection, Semantic/Instance Segmentation
- Multi‑Object Tracking, OCR, Pose Estimation and Action/Activity Recognition
- Facial Detection, Embeddings, Similarity Matching and Anomaly Detection
- Image Enhancement, Video Analysis, Temporal Modelling and Real‑Time Inference
FRAMEWORKS & LIBRARIES
- Python, PyTorch/TensorFlow, OpenCV
- NumPy, Pandas and scikit‑learn
- YOLO, RT‑DETR, Faster R‑CNN or equivalent
- ByteTrack, DeepSORT, MediaPipe or equivalent
- PaddleOCR, EasyOCR, Tesseract and Hugging Face
- ONNX/ONNX Runtime, TensorRT and Open VINO
- FAISS, pgvector or equivalent vector‑search technologies
VIDEO & EDGE AI
- RTSP, ONVIF, FFmpeg/GStreamer
- Video Decoding, Frame Extraction, Codecs, FPS, Bitrate and Preprocessing
- Real‑time Video Processing, Buffering, Reconnection and Pipeline Health Checks
- NVIDIA GPU, Edge AI, Quantization, Pruning and Acceleration
- NVIDIA Jetson or equivalent edge hardware is desirable.
CLOUD & MLOPS
- Azure: Azure Machine Learning, Azure AI Services, Azure AI Video Indexer, AKS, Azure Arc, IoT Hub, Event Hubs, Service Bus, Functions, Container Apps, Blob Storage, Azure Monitor, Log Analytics, Application Insights, Key Vault, Entra ID, PostgreSQL, Cosmos DB and Azure SQL
- AWS: SageMaker, EC2, S3, EKS/ECS, Lambda, IoT Greengrass, Kinesis, CloudWatch, Rekognition and IAM
- Docker, Kubernetes, Azure DevOps/GitHub Actions and MLflow
- CI/CD, Experiment Tracking, Dataset/Model Versioning and Model Registry
- Automated Model Testing, Monitoring, Observability and Rollback
- Infrastructure as Code exposure
SOFTWARE ENGINEERING
- Strong Python, OOP and Modular Development
- REST APIs, Microservices and SQL/NoSQL
- Git and Linux/Shell Scripting
- Unit, Integration and Performance Testing
- Debugging, Performance Profiling and Production Troubleshooting
SECURITY & RESPONSIBLE AI
- Secure Image/Video Data Handling, Encryption, Access Control and Secrets Management
- Secure APIs, Data Retention/Deletion and Privacy-by-Design
- Human-in-the-Loop and Responsible AI
- Bias/Fairness, Logging and Auditability
EDUCATION
Bachelors/Masters degree in Computer Science, AI, ML, Data Science, Electronics or related discipline. Relevant Cloud, AI/ML or Computer Vision certifications are advantageous.
EXPERIENCE
- 5-8 years of overall experience in AI/ML.
- 3+ years of focused experience in Computer Vision and Deep Learning.
- Hands‑on experience training, optimizing and deploying Computer Vision models in production.