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

I’m Shruthi Atkuri, an AI/ML engineer focused on building practical computer vision and data-driven AI systems. I enjoy taking end-to-end problems from data preparation to model training, evaluation, and deployment—especially when the solution can measurably reduce manual effort. From fine-tuning YOLO models for object detection and tracking to creating agentic AI and RAG pipelines with tool calling, I work across the full ML lifecycle. I also value reliable engineering practices like secure backends, testing-based model selection, and clear experiment documentation so teams can confidently adopt what we build.…I’m Shruthi Atkuri, an AI/ML engineer focused on building practical computer vision and data-driven AI systems. I enjoy taking end-to-end problems from data preparation to model training, evaluation, and deployment—especially when the solution can measurably reduce manual effort. From fine-tuning YOLO models for object detection and tracking to creating agentic AI and RAG pipelines with tool calling, I work across the full ML lifecycle. I also value reliable engineering practices like secure backends, testing-based model selection, and clear experiment documentation so teams can confidently adopt what we build.

SHRUTHI ATKURI

AI Engineer, Data Scientist, AI Developer, +1





I’m Shruthi Atkuri, an AI/ML engineer focused on building practical computer vision and data-driven AI systems. I enjoy taking end-to-end problems from data preparation to model training, evaluation, and deployment—especially when the solution can measurably reduce manual effort. From fine-tuning YOLO models for object detection and tracking to creating agentic AI and RAG pipelines with tool calling, I work across the full ML lifecycle. I also value reliable engineering practices like secure backends, testing-based model selection, and clear experiment documentation so teams can confidently adopt what we build.…I’m Shruthi Atkuri, an AI/ML engineer focused on building practical computer vision and data-driven AI systems. I enjoy taking end-to-end problems from data preparation to model training, evaluation, and deployment—especially when the solution can measurably reduce manual effort. From fine-tuning YOLO models for object detection and tracking to creating agentic AI and RAG pipelines with tool calling, I work across the full ML lifecycle. I also value reliable engineering practices like secure backends, testing-based model selection, and clear experiment documentation so teams can confidently adopt what we build.

Available to hire

I’m Shruthi Atkuri, an AI/ML engineer focused on building practical computer vision and data-driven AI systems. I enjoy taking end-to-end problems from data preparation to model training, evaluation, and deployment—especially when the solution can measurably reduce manual effort.

From fine-tuning YOLO models for object detection and tracking to creating agentic AI and RAG pipelines with tool calling, I work across the full ML lifecycle. I also value reliable engineering practices like secure backends, testing-based model selection, and clear experiment documentation so teams can confidently adopt what we build.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Intermediate

Work Experience

AI Engineer (Industry Capstone) at GMDSOFT

March 1, 2024 - June 1, 2024

Developed a computer vision pipeline for a digital forensics setting to detect and track specific objects in CCTV videos. Fine-tuned YOLOv8 and YOLOv7 for object detection and compared them with Faster R-CNN; after team testing, selected YOLOv8 as the final model achieving 95% detection accuracy. Integrated ByteTrack, OpenCV, InceptionV3, and cosine similarity to support object tracking and re-identification. The solution reduced manual video review time by approximately 80%.

AI/ML Engineer Intern at Skyyskill Academy

August 1, 2022 - November 1, 2022

Built and evaluated heart disease classification models using Scikit-learn, Pandas, and NumPy to predict heart disease. Conducted data preprocessing and exploratory data analysis (EDA), then performed hyperparameter tuning to compare different algorithms as part of a team effort. Selected Random Forest as the final model after benchmarking, achieving 83% accuracy. Deployed the model using Streamlit and documented experiments and evaluation results to support the team’s final decision.