Senior GenAI Engineer (AI)

TL;DR

Senior GenAI Engineer (AI): Building and shipping AI agents end to end, including the agent runtime, tools, memory, guardrails, evaluation, and observability with an accent on production-grade multi-agent orchestration and trustworthy tool execution. Focus on designing measurable quality and safety systems, instrumenting per-session/per-turn debugging, and turning working patterns into reusable platform components.

Company

hirify.global is a food marketplace connecting chefs with customers through an AI-driven platform.

What you will do

• Own AI agent features end to end: orchestrator/sub-agent design, tools, system prompts, memory, and the response contract.
• Design and evolve the production agent runtime for low-latency member-facing paths and fast debugging.
• Build trustworthy tool layers (grounded search, retrieval/reranking, cart and account actions) while protecting credentials and identity.
• Implement layered safety: input/output guardrails, intent/clarification handling, refusals, and PII boundaries.
• Advance evaluation: structured checks plus LLM-as-judge with a review queue for disagreements.
• Instrument and platformize: tracing, token/cost visibility, and reusable runtime/eval/guardrail conventions across agent codebases.

Requirements

• Real production experience building with LLMs and agents.
• Strong judgment across grounding and retrieval, tool/context design, memory, cost/latency, safety, and how to verify effectiveness.
• Ability to identify reusable patterns from a working agent and avoid over-engineering.
• Strong Python, plus enough breadth across APIs, cloud, and infrastructure-as-code to own delivery from model call to deployment.
• Track record of setting technical direction and improving engineers around you.
• Ability to explain trade-offs clearly to non-technical stakeholders.

Culture & Benefits

• Remote work.
• Paid vacation: 15 days per year from the start date.
• 16 fully paid Argentinean holidays and paid family leave.
• Monthly healthcare stipend and flexible benefits via a prepaid card.
• 5-year sabbatical: 4 weeks paid after 5 years, plus compassionate leave (3–5 days).
• AI-forward workplace with enterprise access to ChatGPT and Claude; personalized English coach.

Hiring process

• Submit an application; recruiters follow up if the profile looks like a fit.

Location: Engineering Latam (Remote)

Salary: Get paid in USD