Help & Supportcareers@gravityer.com
AI Engineer with less than a year in Generative AI, MLOps, and Agentic Systems
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Highly motivated and skilled AI Engineer pursuing a Bachelor of Technology in Computer Science with a strong focus on Machine Learning, Generative AI, and Agentic Systems. Proven ability to design and implement complex AI models, develop RAG agents, and contribute significantly to open-source projects like Keras. Eager to leverage deep learning expertise and problem-solving skills in innovative technical environments.
Indian Institute of Information Technology Vadodara-ICD
Bachelor of Technology · Computer Science
August 1, 2024 – June 30, 2026
Google Open Source
Open Source Contributor
June 1, 2026 – Present
India
IIITV-ICD
Technical Committee Member
May 1, 2026 – Present
India
CAS - Codebase Agent
January 1, 2026 – Present
• Built a system that enables developers to onboard 10x faster by instantly surfacing architecture patterns, dependencies, and code intent by turning any unfamiliar codebase into a queryable knowledge base. • Developed a Retrieval-Augmented Generation (RAG) agent to ingest, index, and query complex codebases. • Engineered an automated ingestion pipeline using vector embeddings to store external repositories in ChromaDB. • Built a FastAPI interface for AI-driven code efficiency suggestions and project architecture identification.
View ProjectKeystroke Biometrics Model
December 1, 2025 – December 1, 2025
• Engineered a keystroke biometrics model for continuous behavioral authentication, capturing timing dynamics to passively verify user identity without passwords or hardware tokens.
Cloud Cleaning and Enhancement Model
October 1, 2025 – Present
• Designed a model which delivers cloud-free satellite imagery through a fully automated 3-stage pipeline, enabling industries like agriculture, urban planning, and disaster response to make high-confidence geospatial decisions without weather delays. • Designed a custom ResNet-18 architecture for precise cloud segmentation, achieving 94% accuracy. • Built a dual-stage pipeline with Gated Partial Convolutions for de-clouding, improving SSIM by 27%. • Enhanced visual resolution using RRDB (ESRGAN) upscaling to restore sharp satellite imagery details.
View ProjectEmotion Recognition using Spectrograms
August 1, 2025 – Present
• Engineered a real-time emotion recognition layer for applications in mental health monitoring, customer experience, and human-computer interaction by predicting human emotions from raw audio without relying on facial data or wearables. • Developed a EfficientNetV2B0 + BiGRU model achieving 73% validation accuracy across six emotion classes. • Processed 12,206 labeled samples into Mel-spectrograms for deep learning-based recognition. • Evaluated multiple architectures including LightCNN and MobileNetV2, optimizing for lightweight performance.
View ProjectHackathon Finalist (Top 10)
HackNagpur 2.0 (National-level Hackathon by GDG)
December 1, 2025 – Present
Cultural Fit Analysis
The candidate's project diversity, ranging from RAG agents to computer vision and audio processing, indicates a broad interest in AI applications. Their open-source contributions to Keras show a willingness to engage with the wider technical community and contribute to large-scale projects. Participation in a national-level hackathon and a technical committee role further demonstrate initiative and collaborative spirit. The academic focus of their experience (still pursuing a bachelor's degree) suggests a strong theoretical foundation and a drive for continuous learning, which is beneficial for a rapidly evolving field like AI. However, the lack of professional industry experience beyond academic projects and open-source contributions might require some adjustment to corporate operational rhythms.
Soft Skills & Operational Fit
The candidate demonstrates strong problem-solving skills through open-source contributions and hackathon participation. Their involvement in a technical committee suggests teamwork and collaboration abilities. The project descriptions are clear and highlight technical achievements, indicating good communication of complex ideas. The academic nature of projects suggests a learning-oriented and research-driven approach, which aligns well with innovative AI roles.