Senior Full-Stack Architect (Backend-Focused)

IntelliPro Group Inc.•San Francisco, CA

2d•$85 - $90•Onsite

About The Position

This role is for a Senior Full-Stack Architect with a backend focus, responsible for architecting scalable, low-latency backend systems, designing optimized database schemas and data pipelines, and developing/managing APIs and microservices. The position involves integrating advanced AI and LLM systems, building backend workflows for AI agents, and translating designs into functional React layouts. Key responsibilities also include implementing AI observability, optimizing performance, and leading by example in using AI-augmented coding workflows. The role requires extensive experience with Python and Node.js frameworks, agentic AI frameworks, high-performance data syncing, and familiarity with corporate internal tooling and risk/workflow domains. Experience with rapid AI iteration tools is also beneficial.

Requirements

7+ Years of Engineering Experience: A deep background in system design, database architecture, and scaling enterprise-grade backend services, with a recent pivot toward full-stack delivery and AI-augmented workflows.
Core Stack Mastery: Deep proficiency in Python (Flask/FastAPI) and Node.js to develop performant service layers, combined with a strong command of React and TypeScript to build production-grade interfaces.
Real-Time & Protocol Expertise: Expert-level knowledge of RESTful APIs, service orchestration, WebSockets, and gRPC-web to ensure data streams perfectly sync between backend nodes and client apps.
State & Data Lifecycle Strategy: Expert implementation of high-throughput databases, caching layers (e.g., Redis), and modern state management strategies (TanStack Query, Zustand, or Redux) to handle heavy data mutations seamlessly.
AI Tooling & LLM Orchestration: Hands-on experience with LLM orchestration frameworks (LangChain, LlamaIndex, Smolagents, DSPy) as well as advanced prompt engineering.
Mastery of the Cursor code editor (indexing, composer, and chat features) to maintain elite velocity.

Nice To Haves

Extensive experience with Python-based frameworks (FastAPI/Flask) and Node.js environments to construct high-performance microservices and tune both relational and non-relational database models.
Demonstrated technical capability in engineering sophisticated AI agent workflows involving Model Context Protocols (MCPs), function calling, and complex patterns for tool interaction, supplemented by expertise in AI monitoring suites like LangSmith or Opik.
Competency in managing intensive data mutations and fine-tuning server-state synchronization to ensure fluid React-based user interfaces without latency or freezing during real-time backend operations.
Previous familiarity with Client’s development resources, internal data streams, and specific frontend libraries such as Fusion.js, BaseUI, and Styletron.
Proficiency in developing automated monitoring platforms, compliance dashboards, or leveraging distributed task engines including Temporal, Cadence, or Celery.
Practical knowledge of utilizing accelerated prototyping environments like v0.dev or Lovable for the swift creation of internal applications.

Responsibilities

Architect scalable, low-latency backend systems capable of handling massive datasets and live risk-data updates.
Design highly optimized database schemas (SQL/NoSQL) and data orchestration pipelines tailored for FRM automation.
Develop and manage RESTful APIs, GraphQL, and high-performance real-time communication layers (WebSockets/gRPC-web).
Implement microservices architecture and Model Context Protocols (MCPs) to securely connect AI agents with internal enterprise data.
Design, integrate, and deploy production-ready LLM systems using providers like OpenAI and Anthropic.
Build sophisticated backend workflows using AI agent frameworks (LangGraph, CrewAI), function calling, and complex tool-use patterns to automate business logic.
Translate high-fidelity Figma designs or rapid prototypes into pixel-perfect, scalable, and secure React layouts.
Manage complex asynchronous workflows using Redux + Sagas or Zustand, while implementing TanStack Query for modern server-state synchronization.
Build resilient AI systems by implementing AI Observability (e.g., LangSmith, Opik), hallucination detection, and retry logic.
Optimize end-to-end performance to ensure frequent real-time updates do not trigger unnecessary UI re-renders, maintaining a fluid user experience even under high data loads.
Lead by example in using AI-augmented coding workflows, leveraging tools like Cursor, Codex, and GitHub Copilot to accelerate cross-stack feature delivery, refactor legacy logic, and eliminate boilerplate.