Vaibhav Jain

I build ML systems for problems that don't have textbook answers — survivorship bias in regression, hidden bridge entities in multi-hop QA, selective forgetting in language models. If it's weird and constrained, I'm interested.

View Projects Resume Blog

Work Experience and Industry Trainings

AI Training Program Trainee

Intel

May 2024

•Completed a specialized AI training program focused on deep learning, computer vision, and model optimization.
•Built and deployed models using Intel OpenVINO toolkit for edge AI applications.
•Worked on hands-on projects involving object detection, classification, and model inference optimization.
•Enhanced understanding of AI hardware acceleration and efficient deployment strategies.

OpenVINODeep LearningComputer VisionModel OptimizationEdge AI

AI Intern (RAG System)

FinvestFx

Jun 2025

•Developed a Retrieval-Augmented Generation (RAG) system to analyze business receipts and payments from Excel files using FAISS and Gemini.
•Optimized token usage through pre-filtration with text embeddings, reducing API costs and enhancing performance.
•Integrated a chat interface with Hugging Face's InferenceClient to support multi-file interactions with history.
•Gained hands-on experience in building efficient, scalable AI solutions during my internship at FinvestFx.

RAGFAISSGeminiText EmbeddingsHugging FaceAIExcel Automation

Full-Stack Development Training

KIIT University - Batch B2

2025

•Currently pursuing intensive full-stack training covering Java Spring Boot, React.js, and Angular.
•Gaining hands-on experience in building RESTful APIs, state-managed frontends, and full-stack applications.
•Collaborating in team projects simulating real-world software development workflows.

JavaSpring BootReactAngularFull-Stack DevelopmentREST APIs

Developer Intern

Acxiom Consulting Pvt. Ltd.

2025

•Working on the SAP Business Technology Platform (BTP) with a focus on cloud-based application development.
•Developing responsive front-end applications using SAP Fiori and UI5 frameworks.
•Building custom UI components and implementing OData services integration for real-time enterprise workflows.
•Collaborating with functional teams to design user-centric interfaces aligned with SAP UX guidelines.
•Gaining hands-on experience in deployment, versioning, and lifecycle management of SAP BTP applications.

SAP BTPSAP FioriSAP UI5ODataCloud Application Development

Freelance AI Competitor

Shipd AI by Datacurve

2026 - Present

•Participating in competitive machine learning challenges, consistently beating established AI baselines to qualify for global leaderboards.
•Developing and optimizing state-of-the-art models across diverse domains, including Computer Vision (YOLOv8, CLIP), NLP (Mistral, XLM-RoBERTa), and Tabular Data processing (LightGBM, XGBoost).
•Implementing advanced deep learning techniques such as LLM fine-tuning via LoRA, multi-hop reasoning, and stacking ensembles under strict computational constraints.
•Engineering custom features and specialized loss functions for extreme low-data regimes and complex algorithmic tasks like code golf and anomaly attribution.
•Achieved a best rank of #2 on the platform while participating in 16 competitive AI challenges across NLP, computer vision, fine-tuning, and tabular modeling.

Machine LearningLLM Fine-tuningComputer VisionNLPEnsemble ModelingPython

Achievements & Competitions

Gemini API Developer Competition - Project Submission

Google Gemini API Developer Competition 2023

2023 • Online • Global

2023

Developed an end-to-end AI health assistant using Gemini 1.5 APIs
Implemented multimodal input support including text, images and nutritional data

Bajaj Finserv HackRx6.0

AI Hackathon

2025 • Online • Banglore • PAN India

2025

Developed a multimodal RAG bot supporting documents, URLs, images, and web scraping
Designed a custom embedding + vector search pipeline for fast and relevant retrieval

Amazon ML 2024

AI Hackathon

2024 • Online

2024

Participated in Amazon flagship ML hackathon focused on product attribute prediction.
Worked with a multimodal dataset containing product images and textual descriptions.

Amazon ML 2025

AI Hackathon

2025 • Online

2025

Participated in Amazons flagship ML hackathon focused on price prediction.
Worked with a multimodal dataset combining product images and textual descriptions.