Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I'm Kowshika M, an AI/ML Engineer with over 4 years of experience building, fine-tuning, and deploying ML models and AI microservices across financial services, healthcare, and e-commerce. I enjoy translating research into production-ready solutions, leveraging frameworks like PyTorch, TensorFlow, and NeMo Curator to boost performance and reliability. I specialize in scalable cloud deployments, CI/CD, and responsible AI practices. My work spans model optimization with TensorRT/Triton, real-time inference, edge deployment, and bias audits to ensure ethical outcomes across regulated industries. I thrive in cross-functional teams and love turning data into business impact.…I'm Kowshika M, an AI/ML Engineer with over 4 years of experience building, fine-tuning, and deploying ML models and AI microservices across financial services, healthcare, and e-commerce. I enjoy translating research into production-ready solutions, leveraging frameworks like PyTorch, TensorFlow, and NeMo Curator to boost performance and reliability. I specialize in scalable cloud deployments, CI/CD, and responsible AI practices. My work spans model optimization with TensorRT/Triton, real-time inference, edge deployment, and bias audits to ensure ethical outcomes across regulated industries. I thrive in cross-functional teams and love turning data into business impact.

Kowshika M

AI Engineer, Data Scientist, AI Developer, +2





I'm Kowshika M, an AI/ML Engineer with over 4 years of experience building, fine-tuning, and deploying ML models and AI microservices across financial services, healthcare, and e-commerce. I enjoy translating research into production-ready solutions, leveraging frameworks like PyTorch, TensorFlow, and NeMo Curator to boost performance and reliability. I specialize in scalable cloud deployments, CI/CD, and responsible AI practices. My work spans model optimization with TensorRT/Triton, real-time inference, edge deployment, and bias audits to ensure ethical outcomes across regulated industries. I thrive in cross-functional teams and love turning data into business impact.…I'm Kowshika M, an AI/ML Engineer with over 4 years of experience building, fine-tuning, and deploying ML models and AI microservices across financial services, healthcare, and e-commerce. I enjoy translating research into production-ready solutions, leveraging frameworks like PyTorch, TensorFlow, and NeMo Curator to boost performance and reliability. I specialize in scalable cloud deployments, CI/CD, and responsible AI practices. My work spans model optimization with TensorRT/Triton, real-time inference, edge deployment, and bias audits to ensure ethical outcomes across regulated industries. I thrive in cross-functional teams and love turning data into business impact.

Available to hire

I’m Kowshika M, an AI/ML Engineer with over 4 years of experience building, fine-tuning, and deploying ML models and AI microservices across financial services, healthcare, and e-commerce. I enjoy translating research into production-ready solutions, leveraging frameworks like PyTorch, TensorFlow, and NeMo Curator to boost performance and reliability.

I specialize in scalable cloud deployments, CI/CD, and responsible AI practices. My work spans model optimization with TensorRT/Triton, real-time inference, edge deployment, and bias audits to ensure ethical outcomes across regulated industries. I thrive in cross-functional teams and love turning data into business impact.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Language

English

Fluent

Work Experience

Machine Learning Engineer at NVIDIA

November 1, 2024 - Present

Fine-tuned and deployed LLMs using NeMo Curator to boost chatbot performance by 30% in financial services. Developed scalable AI microservices with NIM for real-time fraud detection and automated claim processing in healthcare, reducing model inference time by 40% and increasing throughput by 25%. Optimized inference pipelines with TensorRT and NVIDIA Triton, cutting latency by 50% for e-commerce recommendations and improving conversions. Trained and fine-tuned models with PyTorch and TensorFlow; deployed LLMs as microservices via Docker, Kubernetes and gRPC across AWS. Maintained CI/CD pipelines with GitHub Actions and Jenkins; optimized GPU-based inference on A100/V100. Used MLFlow and DVC for model lifecycle and conducted bias audits for responsible AI in regulated industries.

Machine Learning Engineer at Scale AI

October 1, 2023 - October 1, 2024

Developed AI safety models evaluating LLMs for national security contexts, ensuring cybersecurity and ethical AI standards. Built a model evaluation framework (CyberSec Eval) reducing insecure code risk by 25% and enhancing robustness. Enhanced content moderation by fine-tuning Llama Guard, increasing detection accuracy by 30% across diverse outputs. Implemented real-time edge deployment (Llama Guard 3-1B-INT4) for mobile/edge devices and led Prompt Guard input filtering to prevent unsafe prompts. Integrated PyTorch, TensorFlow, Keras, ONNX, and Docker; established CI/CD with GitLab, Jenkins, and MLflow. Applied transformers, RL, and semi-supervised learning to improve accuracy and reduce false positives.

Software Engineer at Accenture

February 1, 2021 - February 1, 2023

Supported AI-driven predictive maintenance initiatives for BPCL, delivering a 40% reduction in unplanned downtime. Integrated IoT sensors and real-time data collection across 18,000 retail outlets and 25,000 tank trucks, enabling actionable insights for model training. Built data pipelines with Apache Spark and Google Cloud IoT to enable real-time processing, improving operational efficiency by 30%. Analyzed operational data with AI for equipment failure detection, increasing predictive accuracy by 25% and reducing costs; aided automation of BPCL’s workflows through cloud+ML integration, improving supply chain efficiency by 35%.