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

AI/ML Engineer with proven experience building scalable AI systems, LLM-powered microservices, and real-time inference pipelines. Experienced in fine-tuning LLMs, optimizing deep learning with ONNX and TensorRT, and deploying RAG and multimodal solutions. Skilled in FastAPI-based architectures, GPU-accelerated inference, and multilingual NLP applications for Indic languages. Led end-to-end production development from research prototyping to scalable deployment across NLP, deep learning, and multimodal domains.…AI/ML Engineer with proven experience building scalable AI systems, LLM-powered microservices, and real-time inference pipelines. Experienced in fine-tuning LLMs, optimizing deep learning with ONNX and TensorRT, and deploying RAG and multimodal solutions. Skilled in FastAPI-based architectures, GPU-accelerated inference, and multilingual NLP applications for Indic languages. Led end-to-end production development from research prototyping to scalable deployment across NLP, deep learning, and multimodal domains.

Pratik Sonune

AI Engineer, Data Scientist, AI Developer, +2





AI/ML Engineer with proven experience building scalable AI systems, LLM-powered microservices, and real-time inference pipelines. Experienced in fine-tuning LLMs, optimizing deep learning with ONNX and TensorRT, and deploying RAG and multimodal solutions. Skilled in FastAPI-based architectures, GPU-accelerated inference, and multilingual NLP applications for Indic languages. Led end-to-end production development from research prototyping to scalable deployment across NLP, deep learning, and multimodal domains.…AI/ML Engineer with proven experience building scalable AI systems, LLM-powered microservices, and real-time inference pipelines. Experienced in fine-tuning LLMs, optimizing deep learning with ONNX and TensorRT, and deploying RAG and multimodal solutions. Skilled in FastAPI-based architectures, GPU-accelerated inference, and multilingual NLP applications for Indic languages. Led end-to-end production development from research prototyping to scalable deployment across NLP, deep learning, and multimodal domains.

Available to hire

AI/ML Engineer with proven experience building scalable AI systems, LLM-powered microservices, and real-time inference pipelines. Experienced in fine-tuning LLMs, optimizing deep learning with ONNX and TensorRT, and deploying RAG and multimodal solutions.

Skilled in FastAPI-based architectures, GPU-accelerated inference, and multilingual NLP applications for Indic languages. Led end-to-end production development from research prototyping to scalable deployment across NLP, deep learning, and multimodal domains.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Intermediate

Intermediate

Intermediate

Work Experience

Associate AI/ML Developer at Immverse Innovations Pvt. Ltd.

December 1, 2023 - April 1, 2026

Fine-tuned LLMs including LLaMA and Phi-4 on domain-specific multimodal datasets to improve contextual accuracy and reduce hallucinations for enterprise use cases. Engineered a 30–40% latency reduction for the MuseTalk talking-head model by converting the inference pipeline to ONNX Runtime for smoother real-time video generation. Designed and deployed FastAPI-based REST microservices for quiz generation, analytics, and dashboards with modular, scalable integration. Led end-to-end AI/ML solution development across deep learning, NLP, and multimodal domains, from research prototyping to production deployment. Collaborated with cross-functional teams to integrate AI-driven features into production workflows, improving overall user experience and product adoption.

Data Analyst at Stealth Startup

August 1, 2022 - December 1, 2023

Analyzed large-scale financial datasets to uncover trends, anomalies, and actionable insights that informed business strategy and investment decisions. Built and deployed predictive ML models for financial forecasting, improving revenue projection accuracy across quarterly planning cycles. Developed time-series forecasting pipelines using ARIMA, Prophet, and LSTM-based models, reducing forecast error by ~20%. Built automated reporting dashboards in Python and SQL, cutting manual reporting time by ~40% and improving accessibility for non-technical stakeholders. Performed data cleaning, feature engineering, and exploratory analysis on structured financial data to ensure model reliability across multi-source datasets.