🌞 GSoC 2026: Interest in Project #2 - Deep Search AI Assistant on Multimodal Personal Database

[pravallika-mutyala]

Hello OpenVINO Mentors (@adrianboguszewski and team),

I am writing to express my strong interest in Project #2: OpenVINO Deep Search AI Assistant on Multimodal Personal Database for AI PC.

I am a developer with experience in building end-to-end AI applications, specifically focusing on the intersection of Vision, Language, and Audio. My background aligns closely with the goals of this project:

Multimodal Fusion: I developed multisense-ai, a Python-based web application that fuses text, image, and audio inputs to generate context-aware responses.

LLM Fine-tuning & Optimization: I have worked with LoRA (Low-Rank Adaptation) to fine-tune LLMs for specific tasks like financial sentiment analysis on earnings transcripts.

Computer Vision: My project AdaptiveNet-PlantVision utilizes ResNet50 and Grad-CAM for high-precision image classification and interpretability.

I am particularly excited about this project because it moves multimodal AI from the cloud to the AI PC using OpenVINO. I am currently:

Setting up the OpenVINO development environment.

Exploring the openvino_genai API to understand how to integrate local RAG with multimodal embeddings.

Looking for a Prerequisite Task on the Good First Issues board to demonstrate my coding standards.

I would appreciate any guidance or specific tasks the mentors suggest for a candidate with my profile.

Best regards,
Pravallika Mutyala
https://github.com/pravallika-mutyala/