Nasir I.

I ship production code faster using Claude Code as my primary development tool. Every project I deliver, from backend APIs to full-stack web apps, is built through an AI-native workflow (Claude Code Max Plan, GitHub Copilot, Warp AI) that lets me move at the pace of a small team while maintaining 80%+ test coverage and CI/CD from day one.

Currently building AI-powered services for two Fortune 500 companies: an actuarial data processing API (FastAPI, Azure DevOps) for a global reinsurance firm, and cloud infrastructure for a major insurer's 10,000-employee desktop migration (Azure VDI, Terraform).

WHAT I DELIVER

AI Agents & LLM Systems: Built a multi-scope contextual memory platform for autonomous AI sales agents using FastAPI, AWS Bedrock, OpenSearch Serverless, and LangChain. RAG pipelines, vector search, prompt engineering, embeddings, and deploying LLMs to production with proper error handling and monitoring. OpenAI, Claude API, Deepgram, and AWS Bedrock integration. I build agents that persist memory, learn user style, and operate autonomously.

Full-Stack AI Development: MERN stack, Spring Boot (Java 21), Python (FastAPI, Django), Next.js, React Native. From database schema to deployed frontend, I own the entire stack so you coordinate with one person instead of three. Built and maintain FocusHive (focushive.app), a live multi-service platform with Spring Boot microservices and a Next.js frontend. 1,376+ tests across my Stripe integration toolkit alone.

Claude Code & AI-Native Workflow: Claude Code Max Plan is my daily driver for architecture decisions, complex refactoring, TDD, and debugging. Combined with GitHub Copilot and Warp AI, I deliver at a velocity that typically requires a 2-3 person team. This isn't a gimmick; it's how I maintain quality at speed across 59 delivered contracts.

DevOps & Cloud: Azure DevOps pipelines, Docker, GitHub Actions, Terraform. Designed hub-spoke Azure infrastructure with ExpressRoute, Firewall rules, and multi-region DR for an enterprise migration. Multi-environment deployments with monitoring and automated rollbacks.

MLOps: Production NLP pipeline processing 1,000+ daily messages with OpenAI + VADER fallback, delivering 60% reduction in manual moderation and 70% cost savings. Experience with AWS Bedrock inference, embedding pipelines, vector database management (OpenSearch, Pinecone, Chroma), and model monitoring.

BEST FIT FOR

• Teams that want an AI-native developer who ships fast with Claude Code