Instrument and monitor agent behavior across success, failure, latency, cost, and human overrides, with exposure to agent-to-tool payment flows as the ecosystem evolves.
- Collaborate with Product & Design to translate user problems into shippable agent workflows, iterating quickly based on real-world feedback.
- What You’ll Bring:
- Strong experience building LLM-powered systems in production, including prompt engineering, structured outputs, tool/function calling, and agentic coding workflows where agents write, modify, and validate code within guardrails.
Solid backend engineering skills with Python and/or Node.js, designing APIs, async services, and background workers that support long-running, stateful agent workflows.
Experience designing multi-step, agent-driven execution loops (plan → act → evaluate → retry), with explicit state management, idempotency, and failure recovery.
Comfort with distributed systems primitives (queues, pub/sub, durable state, rate limiting, concurrency controls) required to coordinate agents at scale.
Proven ability to instrument and observe agent behavior, emitting structured events for task success, failures, retries, overrides, and code-generation outcomes.
Experience deploying and operating systems on cloud infrastructure (AWS or GCP), using Docker and CI/CD pipelines.
Familiarity with inter-agent communication patterns, including message schemas, shared memory, blackboard systems, or event-driven coordination.
Experience implementing agent evaluation pipelines (prompt regression tests, simulation-based evals, automated scoring, human-in-the-loop review).
Understanding of cost-aware execution for agents (dynamic model routing, caching, batching, budget constraints).
Experience with RAG architectures, vector databases, and persistent memory strategies.
- Exposure to emerging agent protocols (e.g., x402 payment flows, MCP), with the ability to implement evolving specs pragmatically.
- Who You Are:
- You operate with high urgency, agency, and a bias for action, translating ideas into working systems quickly, learning from real-world signals, and iterating continuously.
You pair speed with rigor, bringing strong engineering fundamentals (systems thinking, evaluation discipline, reliability) so each iteration is meaningfully better, not just different.
You embrace a “long is wrong” mindset, simplifying, automating, and removing bottlenecks to consistently reduce cycle time.
You think and act like an owner, making sound tradeoffs and taking end-to-end accountability for performance, reliability, cost, and user experience.
You optimize for technological leverage, building reusable primitives, tooling, and patterns that compound team output over time.
You are growth-minded, working backward from LTV/CAC and prioritizing engineering work that moves activation, retention, conversion, and cost per successful outcome.
- You excel at constraint removal, unblocking execution and increasing innovation per unit time, including through agentic tool use with clear guardrails and observability.
- What We Have To Offer:
- Competitive salary + bonus + equity
Generous PTO + 11 company holidays
Open sick time
100% covered Medical, Dental, Vision for employees
401k with match
Health & Dependent Care Flex Spending Account
Paid professional development
Leadership & growth opportunities
Virtual company and team building events