Ishan Singh Sengar — Machine Learning Engineer

I am a seasoned Machine Learning Engineer 🧠 with over 7 years of experience building scalable AI systems that solve complex, real-world problems. My career is defined by a transition from traditional software engineering to architecting sophisticated ML frameworks and deep learning models at the intersection of research and production. 🚀 Currently, I am contributing to the AI ecosystem at Meta, focusing on two core pillars of ML excellence: Real-Time Multimodal Systems 🎧: I architect robust, fault-tolerant deep learning systems for audio and speech processing. By implementing advanced cross-attention mechanisms and spatially-aware embeddings, I develop models that maintain high performance even under sensor failure—ensuring production-grade reliability for wearable technology. 🕶️ Scalable ML Infrastructure 🏗️: I design and maintain distributed PyTorch training frameworks that standardize pipelines for large-scale research. My work includes upstreaming optimizations to open-source libraries like TorchTNT, focusing on distributed metrics synchronization, multi-GPU efficiency (NCCL), and ensuring the integrity of large-scale model checkpoints. 📊 Core Expertise 🛠️- Deep Learning & NLP: Architecting Transformers, RNNs, and LSTMs for sequence modeling and speech enhancement. 🗣️ Computer Vision: Implementing custom object detection (YOLO), semantic segmentation (U-Net), and sophisticated OCR pipelines. 👁️ MLOps & Tooling: Developing distributed training frameworks using PyTorch DDP and automated AI-assisted coding agents to accelerate the SDLC. ⚡ Data Science at Scale: Utilizing clustering and regression models to drive business intelligence and optimize user engagement. 📈 Technical Toolkit 🧰- Frameworks: PyTorch, TensorFlow, Keras, TorchTNT Languages & Tools: Python, SQL, REST Microservices, AWS, OpenCV, Hydra Distributed Systems: Multi-GPU training, NCCL, Distributed Checkpointing ⛓️ I am passionate about pushing the boundaries of what is possible in AI—from contributing to the Meta-PyTorch ecosystem to architecting high-performance ML pipelines. 🌌 Let’s connect to discuss the future of AI, distributed training, or innovative ML applications! 🤝✨

Stackforce AI infers this person is a Machine Learning Engineer specializing in scalable AI systems for B2B applications.

Location: Indore, Madhya Pradesh, India

Experience: 7 yrs 1 mo