Nov 2024 – Present · 1 yr 6 mos · Hyderabad, Telangana, India · Remote

• Working on production-grade AI and machine learning systems across data preprocessing, model training, evaluation, inference, and deployment, with a focus on scalability, reliability, and cost efficiency.
• Key Contributions:
• Optimized NLP model inference pipelines for production APIs, reducing latency and improving system stability through quantization, batching, and runtime-level optimizations.
• Contributed to a computer vision-based quality assessment pipeline integrated into large-scale data processing workflows.
• Fine-tuned and deployed transformer-based text classification models on domain-specific datasets, validating improvements through offline evaluation and staged production rollouts.
• Built internal evaluation and monitoring tools to track model performance, data drift, and regression issues across production deployments.
• Developed a model evaluation framework tracking 15+ metrics across classification, regression, and ranking tasks, detecting prediction drift within 48 hours using statistical tests and triggering automated rollback workflows.
• Collaborated with data and platform teams to optimize data pipelines, reduce processing bottlenecks, and improve end-to-end system reliability
• Implemented an A/B testing framework for model deployments, measuring business impact across multiple production experiments with statistical power analysis and guardrail metrics.