Aniketh bhosale

I’m a B.Tech graduate in Electronics and Computer Engineering, and I focus on building AI and data-driven systems. My work spans machine learning, deep learning, computer vision, data engineering, and generative AI, with a strong emphasis on turning models into usable products. I’m comfortable moving between experimentation and production-minded design, using tools like Python, SQL, PySpark, Databricks, and modern ML frameworks.

Recently, I completed a Software Engineering Internship (IICS) with Capgemini, where I built hands-on skills in Oracle SQL, PL/SQL, Unix, and Informatica PowerCenter alongside Informatica IICS. I developed ETL mappings and strengthened my understanding of data integration, transformation, and data warehousing concepts. This experience shaped my approach to reliability, data quality, and clear end-to-end pipelines.

In my projects, I’ve shipped end-to-end solutions that blend modeling and engineering. For example, I developed a Pathological Voice Classification model using a hybrid CNN-BiLSTM architecture with data augmentation and 5-fold cross-validation, reaching 87.27% accuracy and 0.939 AUC with CQCC features. I also built an IoT + ML traffic optimization system using Raspberry Pi and Flask for real-time capture and cloud processing, and I created an agentic multilingual loan advisory chatbot using Sarvam AI for translation and OpenAI GPT with RAG-based contextual response generation.

I also prioritize scalable, fault-tolerant data workflows. I built a Dockerized ETL pipeline with Dagster that fetches real-time stock data from the Alpha Vantage API, implementing rate limiting, retry logic, and validation, and persisting results in PostgreSQL using UPSERT. Beyond building, I continuously validate my direction through certifications and collaborative hackathons, including hands-on leadership roles in developer communities.