Specializing in Artificial Intelligence and Analytics. I thrive at the intersection of Generative AI, RAG architectures, and Model Context Protocols (MCP) & Deep Learning.
I have significant experience in rapid prototyping, quickly transforming ideas into testable solutions, developing proof-of-concepts, and MVPs to validate ideas.
Beyond technical implementation, I excel in communication and presentation, bridging the gap between complex backend architectures and impactful user experiences.
| Category | Skills |
|---|---|
| Languages | Python, Java, SQL, Hindi (Native), English (Professional), Korean (Hangul Literacy) |
| AI / Machine Learning | Generative AI, RAG, Model Context Protocol (MCP), Deep Learning, NLP, TensorFlow, Keras, Scikit-learn |
| Frameworks & Tools | FastAPI, React.js, Streamlit, OpenCV, Git, PostgreSQL, Salesforce Cloud |
-
Protly: Protly is an end-to-end machine learning web platform that translates amino acid sequences into high-fidelity 3D protein models using Meta's ESM-2 language model.
-
Prospera: An intelligent financial agent leveraging Gemini 2.5 Flash and the Model Context Protocol (MCP) to provide actionable insights across diverse investment portfolios.
-
Emerge: A hackathon-winning full-stack platform featuring an AI Career Coach and automated, interest-based course recommendations via YouTube Data API.
-
Talent Scout: An NLP-driven interviewing engine that uses real-time response analysis to generate dynamic, expertise-level follow-up questions.
-
MCP RAG: A high-precision Retrieval-Augmented Generation tool utilizing GroundX and MCP to bridge private documents with Gemini AI.
-
Intern | Deutsche Telekom Digital Labs (June 2026 - Present)
- Engineered a dynamic generation pipeline for a production AI chatbot, moving beyond standard text-based chat to render interactive UI components using Agent-to-User Interface (A2UI) framework.
- Integrated a Small Language Model (SLM) fallback mechanism to ensure seamless user experiences and high availability.
- Developed robust parsing capabilities to translate diverse system inputs (ranging from unstructured text to complex JSON) into actionable front-end components.
-
Intern | Sopra Steria (June - August 2025)
- Developed a streamlined RAG tool connecting private documents to Google Gemini AI.
- Utilized GroundX for automated semantic retrieval and high-precision storage of complex file formats.
- Implemented Model Context Protocol (MCP) to allow AI agents to plug into projects without custom glue code.
-
Trainee | Astrea IT Services (June - July 2025)
- Completed intensive training on the Salesforce Cloud Computing platform, focusing on architecture and administration.
- Neural Networks and Deep Learning – DeepLearning.AI.
- Google Data Analytics Specialization – Google.
- AI Algorithm Development with Python – Intel.
- LinkedIn: Rudraksh Mahajan
- Email: rudraksh.r.mahajan@gmail.com