Senior Lead, AI Engineering, GDBS

Senior Lead, AI Engineering, GDBS

South Pole

Multiple Locations

Job summary:

South Pole is looking for an AI Developer to help build the intelligent layer behind our products. You will design, build, and ship production-grade AI features: LLM-powered workflows, retrieval pipelines, and agentic tooling that turn large, messy environmental and carbon-market data into reliable answers.

This is a hands-on engineering role pitched at a medior-to-senior level. We are open to a strong medior developer ready to step up, as well as a senior who can take a vague requirement and turn it into a tested, well-reasoned feature. What matters is that you understand the why behind your choices - not just how to call an API, but when not to, and how to keep an AI system trustworthy, evaluated, and cost-aware in production.

Key Responsibilities:

• AI Feature Development: Design and ship LLM-powered features end to end - retrieval-augmented generation (RAG), structured extraction, summarisation, and agentic/tool-using workflows;
• Pipelines & Integration: Build robust data and inference pipelines that connect our AI layer to product back-ends, vector stores, and source systems (AQ, Salesforce, Google Workspace);
• Evaluation & Reliability: Stand up evaluation, monitoring, and guardrails so AI features are measurably accurate, safe, and stable - not just impressive in a demo;
• Performance & Cost: Optimise latency, token usage, and throughput; apply caching, batching, and model-routing to keep large workloads fast and economical;
• Quality Assurance: Champion rigorous testing and pull-request discipline to prevent regressions and keep deployments stable;
• Collaboration: Partner with the Tech Lead and product owners to translate business problems into AI solutions, and flag where a non-AI approach is the better call.

Technical Requirements - Core Stack

AI & Machine Learning:

• LLM Engineering: Hands-on experience building with LLM APIs (e.g. Anthropic, OpenAI, or Vertex AI/Gemini) - prompt design, function/tool calling, and structured outputs;
• RAG & Retrieval: Practical knowledge of embeddings, chunking, and vector search (pgvector, or a managed vector DB);
• Frameworks: Familiarity with at least one orchestration framework (LangChain, LlamaIndex, or equivalent), and comfort working without one when it's simpler;
• Evaluation: Awareness of LLM evaluation, hallucination mitigation, and basic red-teaming of AI outputs.

Back-end & Data:

• Back-end: Strong Python (FastAPI/Flask/Django), with clean, testable, production-oriented code;
• Database: Solid PostgreSQL - comfortable with relations, indices, constraints, and transactions beyond basic ORM usage;
• Data: Experience handling large, high-volume datasets and streaming / batch processing.

Systems & Cloud-Native Architecture:

• Cloud: Experience with GCP (Vertex AI, Cloud Run, Pub/Sub) or a comparable cloud platform;
• Event-Driven Design: Proficiency with queues and Pub/Sub for asynchronous, event-driven workflows;
• DevOps Fundamentals: Working knowledge of CI/CD workflows and containerisation (Docker; Kubernetes a plus);
• Security: Sound practices around authentication, authorisation, throttling, and handling of sensitive and proprietary data in AI workflows.Systems & Cloud-Native Architecture;
• Cloud: Experience with GCP (Vertex AI, Cloud Run, Pub/Sub) or a comparable cloud platform;
• Event-Driven Design: Proficiency with queues and Pub/Sub for asynchronous, event-driven workflows;
• DevOps Fundamentals: Working knowledge of CI/CD workflows and containerisation (Docker; Kubernetes a plus);
• Security: Sound practices around authentication, authorisation, throttling, and handling of sensitive and proprietary data in AI workflows.

Communication Skills:

• Proficiency in business English;
• Ability to translate complex technical and AI concepts into understandable terms for non-technical business colleagues.

Desirable Requirements:

• Experience with MLOps/model lifecycle tooling, or fine-tuning;
• Exposure to front-end work (React/Next.js) for AI-facing interfaces;
• Interest in climate, carbon markets, or sustainability data.

The Ideal Candidate Profile:

• Proactive Communicator: You are proactive and communicative. In interviews you prefer a conversational approach - walking us through your reasoning and the trade-offs behind your tech choices, rather than just providing the “correct” answer;
• Pragmatic Problem Solver: You don't just make a model work once; you find the root cause, build for reliability, and know when a simpler, non-AI solution is the right one;
• Independent: You can take a high-level requirement and turn it into a fully functional, tested, evaluated feature with minimal oversight;
• Responsible by Default: You care about accuracy, safety, cost, and the trustworthiness of what you ship - a natural fit for a climate company where credibility matters.