Platform AI Reliability Engineer — Remote & Flexible

Softwaremind; Buenos Aires; Híbrido; ARS 105.515.000 - 165.810.000; Jornada completa

Ventajas ofrecidas por este puesto de trabajo

Educational resources

Flexible schedule

Work From Anywhere

Referral Program

Supportive atmosphere

Descripción de la vacante

Softwaremind is seeking a backend engineer to harden an AI copilot for process engineers in oil refineries and chemical plants. You will own the trust layer and grounding mechanisms within a large Python codebase, ensuring the agent says "I don't know" and surfaces assumptions for correction.

You will diagnose misbehavior, extend evaluation harnesses, improve latency, and apply concurrency fixes, all while supporting SOC-like reliability and staged rollouts across teams in

Formación

• Strong Python in large, shared backend codebases.
• Shipped an LLM-based feature to production and built an evaluation that changed a real decision.
• Production debugging from symptom to confirmed root cause.
• Prompt work treated as engineering with measured adherence and regression testing.
• Experience with feature flags and staged rollouts across time zones.

Responsabilidades

• Investigate agent misbehavior from real customer plants and diagnose, fix, and regression-test.
• Build and extend evaluation harnesses that gate prompt changes and model migrations.
• Add reliability mechanisms: assumption surfacing, grounding checks, output-quality reviewers.
• Diagnose and resolve cross-cutting performance and concurrency issues.
• Raise code quality within the team's review conventions.

Conocimientos

Strong Python

LLM in production

Prompt engineering

Observability tooling

Herramientas

Matplotlib

Descripción del empleo

Softwaremind is seeking a backend engineer to harden an AI copilot for process engineers in oil refineries and chemical plants. You will own the trust layer and grounding mechanisms within a large Python codebase, ensuring the agent says "I don't know" and surfaces assumptions for correction.

You will diagnose misbehavior, extend evaluation harnesses, improve latency, and apply concurrency fixes, all while supporting SOC-like reliability and staged rollouts across teams in