Shivaprasad Gowda

Intern

Belagavi, Karnataka, India9 mos experience

AI EnabledAI ML Practitioner

Key Highlights

Developed a hybrid ML framework with R² ≈ 0.99 accuracy.
Engineered a time-series model achieving 92% accuracy.
Designed advanced computer vision systems for logistics.

Stackforce AI infers this person is a skilled AI/ML professional with expertise in energy, agriculture, and logistics sectors.

Contact

Skills

Core Skills

Artificial Intelligence (ai)Data ScienceComputer Vision

Other Skills

Graphics Processing UnitCore MLDeep LearningEvent PlanningCommunicationGenerative AIManagementResearch SkillsInstitutional ResearchNatural Language Processing (NLP)Large Language Models (LLM)Fine TuningBig DataData AnalysisData Mining

About

Passionate about Artificial Intelligence and Machine Learning, I am on a mission to unlock their potential to drive innovation, solve complex challenges, and improve lives. My vision is to leverage technology to create sustainable solutions that contribute to the betterment of society. Fueled by curiosity and a growth mindset, I am dedicated to expanding my expertise and exploring cutting-edge applications of AI/ML. With a focus on continuous learning, I aim to collaborate with like-minded professionals and turn ideas into transformative action. Let’s connect, learn, and build the future together! 🌍✨

Experience

9 mos

Total Experience

9 mos

Average Tenure

9 mos

Current Experience

Indian institute of technology, roorkee

Research Intern

Feb 2026 – Present · 3 mos · Roorkee, Uttarakhand, India · Remote

Computer VisionArtificial Intelligence (AI)Graphics Processing Unit

Indian institute of technology, bombay

AI/ML Intern - FOSSEE Project

Oct 2025 – Feb 2026 · 4 mos · Mumbai, Maharashtra, India · Remote

1.Developed a chemistry-informed hybrid ML framework (PLS + SVR/ANN) to predict crude oil properties (Cetane, Viscosity, PNA, Aniline Point, etc.) using regime-wise residual learning, achieving up to R² ≈ 0.99 (Cetane) under cross-validation.
2.Designed robust preprocessing & forensic data pipeline including TBP-based physics-guided feature engineering, CoDA transformations for compositional constraints (P+N+A=100), ROBPCA outlier detection, and residual-based progressive pruning.
3.Systematically evaluated modeling strategies under small-data constraints (118 assays), demonstrating that data quality and regime-aware modeling outperform naive augmentation and high-complexity models, improving generalization stability across targets.

Artificial Intelligence (AI)Core MLDeep LearningData Science

Tantrafiesta

Event Organizer

Oct 2025 – Oct 2025 · 0 mo · Nagpur, Maharashtra, India · On-site

Event PlanningCommunicationGenerative AIManagement

Gdsc iiitn

AI/ML Core team member

Sep 2025 – Present · 8 mos · Nagpur, Maharashtra, India · On-site

Crispr iiit nagpur

Head of AI Research

Aug 2025 – Present · 9 mos · Nagpur, Maharashtra, India · Hybrid

Artificial Intelligence (AI)

Indian institute of information technology nagpur

Research Intern

May 2025 – Jul 2025 · 2 mos · Nagpur, Maharashtra, India · Remote

Research SkillsInstitutional ResearchComputer VisionNatural Language Processing (NLP)Large Language Models (LLM)Fine Tuning+4

Elevatetrust.ai

Artificial Intelligence Intern

Mar 2025 – Jun 2025 · 3 mos · Pune District, Maharashtra, India · Remote

ML & Computer Vision Intern | ElevateTrust.AI Solutions
[Mar 2025 - Jun 2025]
Spearheaded the development of two impactful AI-driven solutions, showcasing expertise in machine learning and computer vision from concept to deployment:
Agricultural Price Forecasting: Engineered a time-series prediction model using Python and Prophet to forecast agricultural commodity prices, achieving 92% accuracy. Developed an interactive Streamlit dashboard for visualizing trends and enabling data-driven decision-making.
Keywords: Time Series Analysis, Predictive Modeling, Data Visualization, Pandas.
CamEdge - Automated Item Logistics: Designed and built "CamEdge," an advanced computer vision system for real-time detection, object tracking (with ByteTrack), and counting of items (bags/boxes) in dynamic environments like warehouses. The Python-based application, featuring a Streamlit UI, processes both video files and live streams (Webcam/RTSP) using YOLOv8 and OpenCV to automate loaded/unloaded counts within user-defined zones, significantly enhancing operational monitoring capabilities.
Keywords: Object Detection, Real-time Video Processing, Deep Learning, NumPy, UI/UX Design.
This internship provided extensive hands-on experience in the full MLOps lifecycle, including model implementation, data processing, robust algorithm design, and user interface development, while fostering strong collaborative and agile teamwork skills.