Headline

AI @ Transbnk | Ex-EY GDS Technology Consulting | GenAI, Agentic AI, RAG & Multi-Agent Systems | Python, FastAPI, Azure

About

I build AI systems that can reason, retrieve information, interact with tools, and automate complex workflows. Over the last few years, I've worked across Agentic AI, Multi-Agent Systems, RAG architectures, workflow automation, and full-stack development. Most recently, I worked at EY GDS, where I developed multi-agent AI workflows using Azure AI Foundry, FastAPI, Power Automate, and MCP-compliant tool integrations. I enjoy taking products from idea to deployment; designing architectures, building APIs, integrating LLMs, and creating user-facing applications. My core interests include: • Agentic AI & Multi-Agent Systems • Generative AI & LLM Applications • Backend Engineering & System Design • Workflow Automation • Full-Stack Product Development Currently looking for opportunities in AI Engineering, Software Engineering, and Applied AI where I can build impactful products and solve challenging technical problems. Let's connect if you're building something interesting! https://rashidubeyportfolio.vercel.app/

Location

India

AI stack

antigravity

claude code

codex

elevenlabs

grok

hugging face

Work experience

Artificial Intelligence Intern at TBX (formerly TransBnk), Jul 2026 - Present · 3 mos, Bengaluru, Karnataka, India

Consulting Intern at EY, Jan 2026 - May 2026 · 5 mos, Hybrid, Built an end-to-end client POC, spanning a multi-agent Python backend, FastAPI REST layer, and React dashboard. Designed a multi-agent orchestration pipeline using Microsoft Agent Framework with two specialized LLM agents (using Azure AI Foundry) coordinated by asyncio semaphores to process 25+ customer accounts in parallel with per-customer error isolation. Engineered a policy-driven recommendation engine reading from YAML-based compliance config to enforce channel eligibility, contact cooldowns, and regulatory constraints across 3 risk tiers before any LLM involvement. Built a Power Automate workflow integrated with SharePoint to automatically trigger and route outreach emails to customers based on AI-generated recommendations, reducing manual intervention by 30% in the collections outreach process. Implemented async batch processing with FastAPI and Azure Blob Storage with non-blocking job execution, live progress polling, and Pydantic-validated LLM outputs with automatic rule-based fallbacks ensuring 100% response reliability. Delivered end-to-end stakeholder demo covering agent architecture, guardrails design, PII masking strategy, latency optimization, and a Phase 2 roadmap.

Software Engineer at Tuleip AI Solutions, Jan 2025 - May 2025 · 5 mos, Designed UI/UX workflows in Figma and translated them into scalable front-end interfaces using React.js, Next.js, and TailwindCSS, improving user engagement by 15%. Executed AI workflow automation pipelines using n8n, orchestrating backend services and data processing pipelines, reducing manual ops and improving dev velocity. Built RAG-based LLM pipelines using Supabase vector storage for real-time document retrieval, improving processing efficiency by 25%. Engineered API-driven architecture, enabling seamless integration across AI, frontend, and backend layers. Optimized component architecture and state management for scalability and maintainability.

Education

Bachelor of Technology - BTech, Computer Science Engineering (with specialisation in Artificial Intelligence and Machine Learning), Vellore Institute of Technology

mahadevi birla shishu vihar

Public links

github: https://github.com/rashidubey03