Senior Python Developer Needed for Autonomous AI Agent / Claw-Style Workflow System

We are looking for a very strong Python developer with real experience building autonomous AI agents and agentic workflow systems. This is not a basic chatbot project. This is not a simple “connect an LLM to a form” project. This is not a quick prototype or demo. This is a mindset change. Building an AI workflow for one task is very different from building an autonomous agent that can oversee a broader business process, monitor things continuously, take actions, manage state, use tools, and escalate when needed. We are looking for someone who understands that difference. The goal is to build autonomous enterprise-style agents that can support real operational workflows. These agents need to be able to monitor processes, understand context, trigger actions, work across tools and APIs, handle failures, and know when to escalate to a human. Strong Python is required. Claw / OpenClaw-style agent experience is absolutely required. Do not apply if you do not have real experience or serious working knowledge with Claw-style autonomous agents. LangGraph experience is helpful, but this role is not only about deterministic LLM workflows. LangGraph is strong for structured workflows and stateful LLM processes, but this project is more focused on the next step: autonomous agents that can monitor broader processes, take action, reason across state, and escalate when human review is needed. Core Requirements: Strong Python development experience Real experience with autonomous AI agents Experience with Claw / OpenClaw-style agent systems Strong understanding of agent loops, tools, state, memory, planning, monitoring, escalation, and guardrails Ability to build production-grade AI workflows, not just demos API integration experience Backend engineering experience Strong architecture and systems-thinking ability Ability to design for reliability, logging, permissions, audit trails, failure handling, and human-in-the-loop escalation Experience with LLM orchestration frameworks Ability to explain technical decisions clearly Helpful Experience: LangGraph OpenAI / Anthropic APIs Multi-agent systems Tool calling Workflow automation Event-driven systems Background jobs and queues FastAPI PostgreSQL AWS or similar cloud infrastructure Enterprise SaaS workflows Monitoring / observability Permissioned tool execution Human approval workflows Security-conscious agent design What We Are Building Toward: We are trying to move beyond one-off AI tasks and into autonomous agents that can help manage larger business workflows. Examples of the type of thinking we care about: An agent that monitors a process over time An agent that notices when something changes An agent that knows what actions it is allowed to take An agent that can use APIs/tools safely An agent that maintains state and context An agent that logs what it did and why An agent that escalates to a human when confidence is low or approval is required An agent that can recover from failures or retry safely An agent that is useful in real operations, not just impressive in a demo What We Are NOT Looking For: Someone who only built basic chatbots Someone who only used no-code AI tools Someone who only copied tutorials Someone who says they know agents but cannot explain the architecture Someone with no Claw / OpenClaw-style experience Someone who cannot handle a live technical screening Someone who is only comfortable building simple prompt chains Someone who ignores reliability, permissions, logging, and failure handling Screening Process: There will be a live phone/video technical screening. You should be prepared to explain: What you have actually built How your agent architecture worked How the agent monitored state How tools/actions were triggered How failures were handled How human escalation worked How you prevented unsafe or incorrect actions Where LangGraph makes sense Where Claw-style autonomous agents make more sense How you would design this for production, not just a demo