Sagar Mohatkar

AI Engineer with 3.6+ years of experience building production-grade AI, Machine Learning, Generative AI, and Time Series Forecasting solutions on AWS. Skilled in Agentic AI, Multi-Agent Systems, Retrieval-Augmented Generation (RAG), LLM applications, NLP, predictive analytics, and time series analysis & forecasting. Proven ability to design scalable AI systems that automate workflows.

Data Scientist at NeoSOFT Private Limited (2025-02 – Present)

• Developed production-grade Generative AI and Agentic AI solutions to perform RCA, enterprise knowledge management, and lead generation using RAG, LLMs, and AWS.
• Designed and deployed a multi-agent query router (LangGraph) with intent classification and agentic SQL retrieval, resolving ~60-70% of order-status queries without manual agent intervention.
• Fine-tuned BERT models for automated review sentiment and department classification (~6K samples/class), cutting manual review triage by ~80% and reducing negative-feedback escalation time from days to near real-time.
• Built retrieval pipelines using the Unstructured library for mixed table/text document parsing, integrating AWS services (Athena, Lambda, S3, SageMaker) for scalable, production-grade inference.
• Collaborated cross-functionally via Confluence and Jira (Agile) to embed AI solutions into existing enterprise knowledge workflows.

ML Engineer at Big V Telecom Pvt. Ltd. (2023-02 – 2025-02)

• Built end-to-end AI/ML solutions for predictive maintenance, renewable energy forecasting, and operational intelligence using NLP, BERT Fine-Tuning, Time Series Analysis & Forecasting, and Predictive Analytics.
• Built a RAG-based HR policy chatbot handling mixed table/text policy documents, cutting query resolution from hours to seconds and reducing repetitive HR query load by ~30-40%.
• Developed a two-step SARIMA + ML hybrid forecasting pipeline (scikit-learn, TensorFlow) for wind power generation, reducing DSM penalty costs by an estimated 15-25% and improving day-ahead scheduling reliability.
• Delivered interactive and static visualizations (Plotly, Matplotlib, Seaborn) to communicate forecasting accuracy and trends to stakeholders.

Bachelor of Engineering (BE) in Electrical Engineering – Nagpur University (RTMNU) (2018)

Diploma in Engineering – Maharashtra State Board of Technical Education (MSBTE) (2015)