Associate, AI Engineer

Pantheon has been at the forefront of private markets investing for more than 40 years, earning a reputation for providing innovative solutions covering the full lifecycle of investments, from primary fund commitments to co-investments and secondary purchases, across private equity, real assets and private credit.

We have partnered with more than 650 clients, including institutional investors of all sizes as well as a growing number of private wealth advisers and investors, with approximately $65bn in discretionary assets under management (as of December 31, 2023).

Leveraging our specialized experience and global team of professionals across Europe, the Americas and Asia, we invest with purpose and lead with expertise to build secure financial futures.

For further details please visit www.pantheon.com

Purpose of the Position

Pantheon is seeking an AI Engineer to contribute to the engineering delivery of production-ready AI capabilities within Pantheon’s AI Product Squad. As the firm transitions from experimentation into sustained, scalable delivery, this role will be responsible for turning validated AI opportunities and product requirements into secure, reliable, observable, and cost-effective AI products and platform capabilities.

Working in a squad-based model alongside the AI Product Owner, this role will provide technical input and hands-on engineering required to design, build, integrate, deploy and operate AI solutions across Pantheon’s workflows. The AI Engineer will work closely with Software Engineering, Data Engineering, the AI Product Owner and Data & Analytics colleagues (including rotational squad members) to ensure AI capabilities are built on trusted data, integrated into real platforms, and operated in line with Pantheon’s governance standards.

This is a hands-on role for an engineer who thrives on applied AI delivery, LLM application architecture, retrieval and agent patterns, evaluation/observability, and production hardening, combined with strong judgement on trade-offs across quality, risk, latency and cost.

Key Responsibilities

Engineering involvement within the AI Product Squad

• Act as a core part of the AI Product Squad, shaping the technical approach to deliver the AI roadmap set by Technology leadership and prioritised by the AI Product Owner.
• Contribute to solution architecture for AI products/capabilities, including design decisions, build plans, technical risks, dependencies, and delivery sequencing.
• Work collaboratively with rotational squad members (Software Engineering, Data Engineering, Data & Analytics, QA), establishing consistent patterns, coding standards, and delivery practices.
• Partner with the AI Product Owner to translate product requirements into delivery-ready technical work items, acceptance criteria, and release plans.

Build and productionise AI capabilities

• Design and implement AI-enabled products and services, including LLM applications, agents, and workflow automations that integrate with Pantheon’s platforms and operating processes.
• Implement core patterns such as retrieval-augmented generation (RAG), tool/function calling, structured outputs, and orchestration, applying the simplest effective approach for each use case.
• Engineer integrations with enterprise data and systems (APIs, document stores, databases), ensuring correct permissions, provenance, and auditability.
• Build reusable components (e.g., retrieval pipelines, prompt/tool schemas, policy guardrails, evaluation harnesses) to accelerate future delivery and consistency.
• Actively and consistently track emerging AI techniques, architectures, and tools beyond current approaches (e.g. RAG, agents), and drive their evaluation and adoption.

Quality, evaluation, and AI observability

• Define and implement evaluation strategies for AI systems, including regression test suites, gold sets, offline/online evaluation, and human review workflows where appropriate.
• Implement AI observability and monitoring to track quality, reliability, behaviour, latency, and cost in production (including evaluation of native platform capabilities and specialist tools).
• Establish feedback loops from monitoring and user outcomes to drive continuous improvement, defect reduction, and drift management.

Production readiness, reliability, and cost management

• Drive non-functional requirements for AI delivery: security, performance, resilience, scalability, supportability, and cost control.
• Implement operational practices (logging, tracing, dashboards, alerting, runbooks, incident readiness) suitable for production environments.
• Manage system-level cost/performance trade-offs (token usage, model selection, caching, retrieval efficiency, latency budgets), and implement controls/guardrails to keep usage predictable.

Governance, security, and responsible AI by design

• Ensure AI solutions comply with Pantheon’s AI governance framework, data protection requirements, and responsible AI principles.
• Design controls including access management, data handling safeguards, transparency, and human-in-the-loop considerations aligned to the risk profile of each product.
• Identify and escalate technical risks (e.g., data leakage, prompt injection vulnerabilities, model behaviour issues), and implement mitigations as part of the engineering lifecycle.

Documentation and handover

• Produce clear engineering documentation including architecture diagrams, configuration guides, test/evaluation reports, and operational runbooks.
• Ensure effective handover and/or shared ownership models with relevant engineering functions for long-term support and lifecycle management.

Knowledge & Experience Required

Core software engineering

• Strong engineering fundamentals, including designing maintainable services/components, writing testable code, and working effectively within modern SDLC practices (CI/CD, code review, version control).
• Experience building and integrating APIs and working with authentication/authorisation concepts.
• Comfort operating production systems, including monitoring, troubleshooting, and improving reliability over time.
• Hands on engineering experience with enterprise AI platforms and tooling (e.g., Databricks/Agent Bricks, ChatGPT Enterprise, developer-focused AI tools such as Claude Code).
• Proficiency in Python, with experience developing production-quality code for data processing, ML, or AI-enabled applications.
• Hands-on experience with machine learning frameworks such as PyTorch and/or TensorFlow, including integrating trained models into applications and services.
• Understanding of modern deep learning architectures, particularly Transformer-based models and attention mechanisms, sufficient to reason about model behaviour, limitations, and engineering trade-offs when building LLM-powered systems.
• Experience building, deploying, or operating services on the Microsoft Azure platform, including the Azure AI Foundry.

Applied AI / LLM engineering experience

• Hands-on experience building AI-enabled products, ideally including generative AI and agent-based patterns.
• Practical experience with one or more of: RAG, vector search/embeddings, prompt and tool schema design, structured outputs, orchestration frameworks, and model routing.
• Strong understanding of common LLM failure modes and mitigations (hallucinations, injection, leakage, overreach, sensitivity/PII concerns).
• Experience using LLM application frameworks such as LangChain, LlamaIndex, Haystack, or equivalent libraries for building and orchestrating LLM-based workflows.
• Hands-on experience with vector databases or retrieval infrastructure, including platforms such as Pinecone or Weaviate, and/or cloud-provider-native vector search solutions, for implementing and operating RAG pipelines at scale.
• Understanding of trade-offs across retrieval approaches (chunking strategies, embedding models, hybrid search, latency, cost, and relevance) in production environments.
• Experience with AI observability/evaluation tooling (e.g., LangSmith and MLflow) and implementing monitoring for quality and cost in production.

Delivery and collaboration

• Experience working in an Agile environment and collaborating with product owners/managers to deliver iteratively.
• Ability to translate ambiguous problem statements into robust technical implementations with clear trade-offs and measurable outcomes.
• Strong communication skills across technical and non-technical stakeholders.

Personal attributes

• High ownership and delivery discipline; comfortable leading technical execution end-to-end.
• Pragmatic, outcomes-oriented, and comfortable working in fast-moving environments with evolving requirements.
• Strong judgement on when to prototype quickly versus when to engineer for production from day one.

Desirable

• Experience in financial services, investment management, or other regulated environments.
• Exposure to data platforms, analytics tooling, and governance processes for using structured and unstructured datasets at scale.

#LI-SB1

This job description is not to be construed as an exhaustive statement of duties, responsibilities, or requirements. You may be required to perform other job-related duties as reasonably requested by your manager.

Pantheon is an Equal Opportunities employer, we are committed to building a diverse and inclusive workforce so if you're excited about this role but your past experience doesn't perfectly align we'd still encourage you to apply.