Agent Platform & Life-Support Engineer

BAM Ventures; New York (NY); On-site; USD 140,000 - 190,000; Full time

Job summary

Aisle in New York is seeking an Agent Builder to join our internal agent stack, currently 8 agents with a plugin ecosystem. You’ll work with our founding agent engineer on production-ready agent capabilities and life-support systems built on OpenClaw.

You’ll design tooling for operators, manage memory and prompts, and drive experiments to improve agent reliability and performance across the stack.

Qualifications

• Hands-on production experience with at least one agent framework (e.g., OpenAI Agents SDK, Claude Code, Codex, Cursor, OpenClaw, LangGraph, Hermes, or similar).
• Familiarity with sessions, context windows, memory systems, and tool/function calling.
• Strong prompt design and context management skills.
• Ability to decompose operator tasks into agent capabilities and design reusable patterns.

Responsibilities

• Partner with domain experts and operators to translate tacit knowledge into agent capabilities.
• Own and improve agent life-support systems: memory, prompts, context windows, retrieval, eval loops, observability.
• Design and build background tools, plugins, and skills for agent expansion.
• Maintain system health in production by diagnosing and fixing root causes, not just symptoms.
• Run experiments to measure impact and refine agent behavior.
• Develop and share reusable patterns across memory, prompting, context, tooling, and orchestration.

Skills

Agent frameworks

Context management

Memory systems

Tool calls

Prompt design

Operator collaboration

Architectural thinking

Experimentation

Debug traces

Decomposition & orchestration

Token minimization

User perspective

Collaboration

Tools

OpenClaw

Hermes

Job description

Aisle in New York is seeking an Agent Builder to join our internal agent stack, currently 8 agents with a plugin ecosystem. You’ll work with our founding agent engineer on production-ready agent capabilities and life-support systems built on OpenClaw.

You’ll design tooling for operators, manage memory and prompts, and drive experiments to improve agent reliability and performance across the stack.