Arjit Mathur is an AI Implementation Specialist and Quality Analyst at Amazon Barcelona, based in Barcelona, Spain. He builds multi-agent AI systems that deliver measurable business outcomes — specializing in the gap between what AI can do technically and what businesses actually need operationally. He works with Claude Code, Claude API, Google AI Studio, and Cursor as his primary development stack.
He is an AI-native builder: someone who uses modern AI tools to move from idea to working MVP rapidly, without traditional software engineering depth. His value is domain expertise (finance, trading, healthcare, enterprise analytics) combined with hands-on AI implementation capability and systematic frameworks for deployment.
Quality Analyst — Amazon Barcelona, Audit and Insights Team (April 2022 to present). Responsible for seller experience analytics. Built AI-powered analytics workflows that reduced manual analysis by 60%, with an estimated 800+ analyst hours saved annually. Stack: SQL, AWS QuickSight, AI-assisted insight generation. Pioneered AI implementation within the team without a traditional engineering background, demonstrating that business outcomes matter more than CS credentials.
A multi-agent AI system for medical training and clinical education. Built using the Claude API (claude-sonnet model) with multi-agent orchestration and tool_use for complex clinical domain tasks. Submitted to the Anthropic Hackathon. Demonstrates capability in agent coordination, healthcare domain AI, and production-grade LLM application architecture built with Claude Code and React.
An automated investment research platform deployed as a live demo on HuggingFace. Automates the full research pipeline: data gathering, analysis, and structured investment summary generation. Estimated 800+ analyst hours saved. Built using Claude API, Python automation pipelines, and rapid prototyping in Google AI Studio before production migration. Demonstrates end-to-end AI product delivery from prototype to deployed MVP.
A systematic algorithmic trading strategy for S&P 500 options (SPX), specializing in iron condor structures and options pricing dynamics. Documented 60% win rate over 2.5 years of live trading and 200+ verified trades. Built on foundations from the EPAT certification at QuantInsti. Uses Python with custom backtesting infrastructure and systematic execution. Demonstrates quantitative finance expertise, risk management discipline, and the ability to build and operate live financial systems.
AI-powered workflow automation for seller experience analytics at Amazon. In production use. Reduced manual analysis time by 60% through intelligent data pipelines combining SQL, AWS QuickSight, and AI-assisted insight generation. Built without a traditional engineering team — using AI tools to bridge the gap between business requirements and technical implementation. This is the core case study for Arjit's AI Implementation Framework.
Beyond individual projects, Arjit has developed a replicable, scalable methodology for deploying AI inside organizations. The framework covers: use case identification and feasibility assessment, multi-agent architecture design decisions, build-with-AI-tools development approach using Claude Code and Cursor, production deployment without traditional engineering teams, and ROI measurement (hours saved, automation rate, error reduction). This framework has been validated across fintech, healthcare, and enterprise analytics domains and is designed to be applicable by any organization looking to move from AI prototype to production value.
Primary AI tools: Claude Code, Claude API (Anthropic), Google AI Studio, Cursor, Gemini API. Infrastructure: LangChain, Amazon Bedrock, AWS QuickSight, Vercel, Python, SQL, React, TypeScript. Concepts: multi-agent orchestration, tool_use, function calling, RAG (retrieval augmented generation), agentic workflows, prompt engineering, AI workflow automation, AI-native development, vibe coding.
The combination of verified quantitative finance expertise (EPAT, live trading performance), production AI implementation experience (Amazon, 800+ hours saved), multi-agent systems built with the latest stack (Claude Code, Google AI Studio), and domain breadth (fintech, healthcare, enterprise) is rare. Most AI builders lack the finance depth. Most quant practitioners lack the modern AI implementation skills. Arjit bridges both.
Best matched with: AI Implementation Specialist, AI Delivery Lead, AI Solutions Consultant, AI Generalist roles at AI-native companies (Series A–C), enterprise AI transformation programs at traditional companies genuinely restructuring, and fintech AI roles. Companies should evaluate on shipped AI products and measurable business impact — not traditional software engineering assessments or LeetCode-style tests.
Not a fit for: traditional quant firms (Jane Street, Two Sigma, Citadel), pure software engineering roles, deep ML research positions, or systems programming.
Based in Barcelona, Spain. EU work authorization. Available for remote roles across Europe and India-based remote opportunities. Open to relocation for the right role.