Lead AI Engineer

About Allexive

Allexive is the AI transformation practice within the Overdose Group. The team helps mid-market and enterprise organisations in New Zealand and Australia move from experimenting with AI to running it as part of day-to-day operations by delivering systems that actually work.

The role sits at the intersection of delivery, client advisory, and pre-sales, supporting the technical direction of the business as it expands into new markets and services.

Workstreams

• Build: Deliver production-ready AI solutions directly into client environments, whether through a proof of concept or a multi-phase build sprint.
• AI platform implementation: Create the foundations for a client’s AI operating model, including the data layer, hosted platform, context architecture, starter libraries of skills and agents, and the change management needed to embed the new way of working. This stream may roll into a monthly retainer.
• Enterprise AI infrastructure: Design the shared AI layer that underpins a client’s broader strategy, including gateway patterns, controls, agentic commerce protocols, vendor evaluation, and phased roadmaps for senior approval.

Role summary

You will be the technical lead across engagements, owning the client-facing technical relationship, helping shape opportunities from a technical standpoint, and guiding delivery with internal engineering capacity and delivery partners. The role requires someone comfortable staying hands-on while also directing others.

Key accountabilities

• Lead technical delivery across build projects, AI platform work, and enterprise AI infrastructure engagements.
• Translate ambiguous problems into working software and direct the engineering effort needed to ship it.
• Own the technical relationship with clients throughout delivery.
• Facilitate working sessions, decision gates, and demos.
• Represent the technical voice in discovery and scoping conversations.
• Assess feasibility, help shape proposals and statements of work, and identify risks or scope issues early.

What the team is looking for

The preferred background is roughly 6 to 10 years in engineering, including about 2 to 4 years working directly with modern AI. The exact number of years matters less than the overall profile.

• You keep close to the fast-moving AI ecosystem and regularly follow the latest model, tooling, and research updates.
• You are willing to experiment with emerging tools and understand where enterprise-grade use begins and ends.
• You have strong command of core AI concepts such as context windows, attention, evaluation design, retrieval approaches, MCP architecture, and agent control flows.
• You have worked in professional services environments with mid-market or enterprise clients and understand billable delivery.
• You build real products and have shipped production AI systems, not just prototypes.
• You communicate clearly with clients, engineers, and executives, including in difficult conversations.
• You understand the commercial side of delivery, including scope management and early risk escalation.
• You are based in New Zealand, ideally in Auckland, and can travel occasionally to client sites.

Technical expectations

Applied AI

• Hands-on production experience with at least one agent framework such as Google ADK, Microsoft Agent Framework, or Claude Agent SDK; experience with two is especially valuable.
• Strong understanding of LLM capabilities and patterns, including tool use, prompt caching, structured outputs, streaming, function schemas, retries, and fallbacks.
• Experience designing context and MCP architectures, including servers, clients, tools, resources, and prompts, and building or extending an MCP server.
• Working knowledge of retrieval-augmented generation patterns, including chunking, embeddings, hybrid search, reranking, and vector databases.
• Ability to build grounded generation workflows, citation approaches, and hallucination controls.

LLMOps

• Experience with automated evaluation methods for non-deterministic AI systems, both offline and online.
• Understanding of observability, tracing, prompt versioning, regression testing, and production monitoring.
• Practical knowledge of safety and red-team practices such as prompt injection defense, handling personal data, moderation, and auditability.
• Ability to optimise production systems for cost and latency.
• Comfort operating the systems that keep a live AI product reliable, not just a demo.

Enterprise AI infrastructure

• Experience thinking about shared AI infrastructure at enterprise scale, including gateway patterns, capability taxonomies, A2A and agentic commerce protocols.
• Familiarity with shared control layers such as identity, delegation, mandate registries, policy engines, and state management.
• Ability to design agents with security, auditability, and a controlled blast radius in mind.
• Comfort discussing vendor selection, gateway placement, and governance boundaries with enterprise architects.

Engineering background

• Strong skills in TypeScript/Node or Python; both is a plus.
• Comfort working across cloud platforms such as AWS, GCP, Azure, and Vercel, with AWS especially valuable for the NZ/Australia market.
• Experience integrating AI systems with complex enterprise data, undocumented APIs, and client software stacks such as CRMs, ERPs, file stores, and communication tools.

Nice to have

• Public technical writing, conference speaking, or open-source contributions.

What you will get

• A senior role in a fast-moving business with active clients and measurable outcomes.
• Direct access to the founder and senior leadership team.
• Exposure to enterprise AI transformation work across New Zealand and Australia.
• Compensation made up of salary plus equity or profit share, to be agreed based on seniority and positioned competitively against senior engineering and consulting roles in New Zealand.
• Remote-friendly work arrangements within New Zealand, with travel to client sites when needed.
• Budget available for tools, models, and learning, with an expectation that the team uses the technologies it recommends.

How the team works

• The company works as an AI-native team, using tools such as Claude Code, agentic workflows, and parallel-agent orchestration to deliver work.
• You will not work alone; senior leaders stay involved in engagements and engineering support is available.
• The team prefers shipping in weeks rather than quarters and focuses on incremental value.
• Consulting theatre is avoided in favour of practical, working deliverables.
• Clients are given direct, honest advice, including when AI is not the right answer.

Location and working style

This role is remote-friendly within New Zealand, with Auckland preferred. Occasional travel to client locations is expected.

Compensation note

Salary and equity or profit share will be negotiated based on seniority. The package is intended to be competitive with senior engineering and consulting roles in New Zealand.