Platform AI Software Engineer

About the role

Software Mind is looking for a Platform AI Software Engineer to own the trust layer of an AI copilot used by process engineers in oil refineries and chemical plants. You will work on a large Python codebase to ensure the agent’s answers are reliable, grounded in real data, and delivered with acceptable latency.

Key responsibilities

• Investigate and fix agent misbehaviour reported from live plant deployments, turning each case into a diagnosis, a fix, and a regression test.
• Build and extend evaluation harnesses that gate prompt changes and model migrations.
• Implement reliability mechanisms such as assumption surfacing, grounding checks, and output‑quality reviewers.
• Diagnose and resolve performance and concurrency issues affecting latency and production stability.

Required profile

• Strong experience with Python in large, shared backend codebases.
• Delivered an LLM‑based feature to production and built evaluations that influenced model or feature decisions.
• Proven ability to debug production issues from symptoms to root cause, including latency spikes and concurrency errors.
• Comfort with feature‑flag discipline, staged rollouts, small PRs, and async collaboration across US and Australia time zones.
• Excellent written English.

Required skills

• Python
• GCP (Vertex AI) – or comparable AWS/Azure services
• Observability tooling (tracing, structured logging, latency monitoring)
• Charting/plotting libraries such as Matplotlib
• Experience with LLM prompt engineering and evaluation frameworks

What we offer

• Educational resources and support for continuous learning
• Flexible schedule and Work‑From‑Anywhere policy
• Referral program
• Supportive and chill atmosphere