Staff AI Engineer

AirTrunk

Location not specified

Full Time

Posted September 28, 2026

Climate Technology

Apply Now

Application opens on company website

Listing evidence and provenance

Source: greenhouse View original listing
Freshness: Open in the current public read path; this is not an employer guarantee.Last observed September 29, 2026
Classification: Modern power-system classification not yet published for this listing.Legacy source industry field: Climate Technology
Pay provenance: Pay not available

Job Description

Working at AirTrunk is a once in a lifetime opportunity to fast-track your career and amplify your impact.

Key Responsibilities

As a Staff AI Engineer at AirTrunk, you will lead the design and delivery of agentic AI applications and the software that brings them into everyday use. This hands-on role combines applied AI, software development and architectural judgement. You will build systems that plan, use tools and carry out multi-step work, taking them from experimentation to reliable production operation. Working with Product, Data, Platform and Security teams, you will shape technical direction, mentor engineers and improve how our teams work.
You’ll design and build AI agents that can plan tasks, use tools and adapt to results. You’ll choose the right agent approach, give agents the context and memory they need, connect them securely to enterprise systems, and test them against realistic tasks to improve their reliability. You’ll also write and review production code, shape application architecture, and work with Product and Data teams to turn business needs into useful software.
You’ll take AI applications from early prototypes through to production and ongoing support. You’ll establish automated tests, agent evaluations and controlled release processes, then monitor task success, speed and cost to identify and resolve issues. Working with Platform engineers, you’ll support deployment and operational readiness across Azure, Databricks and related services, making sure applications can recover safely when interrupted and avoid repeating actions.
You’ll build privacy, security and responsible AI controls into agent design, including appropriate access permissions, protected credentials and human review for consequential actions. You’ll work with Security and business teams to assess risks, test for unsafe or unreliable behaviour, and agree practical safeguards. You’ll also lead complex technical work, make architectural decisions, mentor engineers and share reusable tools and lessons with AirTrunk’s Data and AI community.

Requirements

10+ years’ experience in software engineering, AI/ML engineering or a related discipline, with hands-on delivery and operation of complex production systems, including agentic applications using tools, orchestration and context management.
Strong software architecture skills across APIs, backend services, distributed systems and enterprise integrations, with sound judgement about complexity and trade-offs.
Advanced Python skills and strong engineering fundamentals, including automated testing, code review, version control and maintainable design; working knowledge of SQL and data access patterns.
Practical understanding of context engineering, memory and agent evaluation, including how to assess task completion, tool behaviour and failure recovery.
Experience delivering secure cloud applications using CI/CD, containers and observability, in partnership with platform teams; understanding of application security, privacy and Responsible AI, including identity, authorisation and risks from model outputs and tool execution.
Demonstrated technical leadership, including guiding architectural decisions, planning pragmatic delivery, influencing across teams, mentoring engineers and improving quality through reusable patterns and shared practices.
Experience with Azure and Databricks, or comparable cloud and AI platforms, is beneficial.
Experience with TypeScript or JavaScript and web application development is beneficial, particularly for delivering complete AI user experiences.
Experience with agent SDKs, MCP integrations and AI-assisted software development is beneficial, with strong review and verification practices.