Senior AI/ML Engineer

Lead end-to-end development, optimization, and deployment of Transformer-based NLP models for real-world applications with a focus on language models like BERT, GPT, and T5.

Design and manage scalable, low-latency inference pipelines achieving 10,000+ requests per second using distributed training and serving technologies such as Kubernetes, Triton, and multi-GPU clusters.

Own ML infrastructure including scalable pipelines (feature store, model registry, CI/CD), containerization, microservices API design, and observability for model performance and zero-downtime updates.

7+ years of professional experience in NLP with at least 3 years dedicated to Transformer models (BERT, RoBERTa, GPT).

Proficient in PyTorch, TensorFlow, Hugging Face transformers, and distributed training frameworks such as DDP, DeepSpeed.

Experience deploying low-latency model APIs with Kubernetes, Docker, FastAPI, and inference servers like Triton.

Location requirement: Gurugram, India (Full-Time, Onsite)

Experienced in architecting and operating large-scale, production-grade NLP models and inference pipelines with focus on throughput and latency optimization.

Skilled in distributed systems, hybrid cloud strategies, and container orchestration to build cost-efficient and highly available ML deployments.

Capable of leading technical mentorship and cross-functional collaboration to translate business needs into state-of-the-art NLP solutions with measurable impact.