The Digital Canvas of My Story
About Me
I am Shriyans Naik, a Computer Engineering graduate with a profound passion for algorithms and data structures. My journey began with a Bachelor's degree in Technology, where I built a robust foundation in computer engineering. During this academic exploration, I dived into software engineering, data engineering, data analysis and app development through internships, expanding my skills and perspectives.
While developing a keen understanding of these foundational elements, I discovered a profound interest in the transformative world of generative AI. The challenge of mastering algorithms took a fascinating turn as I delved deeper into the possibilities of artificial intelligence, exploring its potential to create and innovate.
Beyond the screen, I find inspiration in the celestial wonders as an astronomy enthusiast. Whether on a hiking trail, capturing moments through photography, or navigating the complexities of finance to understand investments and economics, I approach each endeavor with the same enthusiasm and dedication.
Let's connect and explore opportunities to collaborate, share insights, or discuss our shared interests. Together, we can aim for the stars, both in the tech world and beyond.
20+
Projects Completed
2+
Years of Experience*
750+
Questions solved on Leetcode
Experience
Data Scientist
@National Stock Exchange
- Streamlined DRHP risk review process by building a domain-filtered vector database of historical risks and implementing semantic similarity matching, reducing review time from 1.5 months to 3 days
Data Scientist
@Jio Platforms Ltd.
- Achieved similar sentiment analysis accuracy to LLaMA 3.1 405B by fine-tuning LLaMA 3.1 8B with QLoRA, reducing GPU utilization by 98%
- Deployed LLaMA 3.3 70B in production using vLLM's continuous batching to serve 128 concurrent requests, enabling support for 10+ applications
- Optimized data ingestion pipelines using Ray to parallelize API calls and embedding generation, reducing processing time from 1 hour and 23 minutes to 12 minutes
- Built an LLM-based summary pipeline using Llama-3.3 70b model, condensing 80-page Oil Market Reports into 6-page insights, saving 480 hours and ₹1.2 million annually
- Automated DevOps ticket validation for Kafka topic creation using Qwen 2.5, reducing ticket closure time from 2 days to 3 hours by ensuring required fields and notifying creators via Teams (Power Automate)
- Developed a Retrieval-Augmented Generation (RAG) pipeline, enabling question answering on over 1,000+ pages of documentation stored on Azure DevOps, improving knowledge retrieval efficiency.
Data Science Intern
@Jio Platforms Ltd.
- Designed CDC pipelines leveraging SQL Record Manager to track Azure DevOps Wiki changes, ingesting over 10,000 wiki updates into vector databases in real-time
Software Engineering Intern
@Jio Platforms Ltd.
- Created a web portal for admins, enabling efficient querying of data for over 500 users simultaneously. Implemented JWT authentication for seamless user authentication and role-based permission assignment. Strengthened security with session management, restricting users to a single active instance at any given time
Data Science Intern
@Jio Platforms Ltd.
- Leveraged Cosine Similarity algorithm to personalize article recommendations based on user reading material, leading to a 15% increase in user retention and a 20% improvement in click-through rates. Enhanced prediction model accuracy by 10% through hyperparameter tuning, minimizing organizational losses
React-Native Developer Intern
@Image Online Pvt. Ltd.
- Created and deployed a cross-platform mobile app with React-Native, slashing development time by 90%
- Utilized Redux to centralize the application's state management, resulting in a 30% reduction in API calls
Education
Bachelor of Technology in Computer Science
CGPA: 9.55- Coursework: Algorithms and Data Structures, Applied Data Science, Big Data Analytics, Object-Oriented Programming, Natural Language Processing, Machine Learning, Artificial Intelligence