See how candidates think and approach the work this role demands, before the phone screen. We'll build a video challenge from this posting, and you can edit or share it before it goes live.
A major bank is standing up a dedicated AI solutions function inside its core control and finance divisions (risk, controllers, compliance, treasury, HR), signaling that generative and applied AI are being pushed into the firm's regulated back-office machinery at VP seniority rather than kept to front-office trading or experiments.
The Core Engineering builds and operates the platforms, applications, data solutions, models, and analytics that power critical processes for The Core divisions of the firm (e.g., Risk, responsible for the risk profile of firm activities; Controllers, responsible for the financial control and reporting obligations; Compliance, responsible for the firm’s compliance, regulatory, and reputational risks; Corporate Treasury, responsible for the firm’s liquidity, funding, balance sheet, etc.; and Human Capital Management, responsible for attracting, developing, and managing a global workforce). A centralized engineering structure in support of The Core enables a common platform model and operating framework that promotes consistent governance and scalable solutions, leveraging cloud, AI, and machine learning for innovation and efficiency. The Core Engineering’s 2,000+ engineers and strats deliver engineering, data, analytics, and quantitative capabilities within six business units:
Summary
The Core Engineering Shared Services AI team operates as an internal forward-deployed AI engineering function, embedding with The Core business teams to identify priority workflows, prototype agentic AI solutions, and productionize cloud-native applications that deliver measurable business impact. This hands-on role combines client-facing problem discovery, rapid engineering execution, production AI architecture, and structured transition to receiving teams that will operate and extend the solutions over time.
As an AI Application Engineer, you will work directly with internal business and engineering partners to convert ambiguous operating problems into secure, reliable AI products. You will define where agents should act, where they should assist, and where human approval or control gates are required, then design and deliver solutions that are adopted in real workflows and can be supported in production.
Key Responsibilities
Rapid Prototyping & Application Delivery: Lead the design, build, deployment, and operationalization of cloud-native AI applications, using modern software engineering practices, CI/CD pipelines, and automated testing to accelerate reliable delivery.
Business Partnership & Solution Architecture: Partner with business and engineering teams to identify high-impact AI opportunities, translate requirements into cloud-optimized architectures, and define scalable data models and technical specifications.
Forward-Deployed AI Engineering: Embed with The Core business teams to map workflows, identify pain points, assess agentic automation opportunities, define control and approval boundaries, iterate with users, and drive adoption through pilots, feedback loops, and measurable outcomes.
Production AI/ML Integration: Build AI applications using LLM APIs, retrieval-augmented generation, embeddings, vector search, prompt and context management, structured outputs, response validation, and evaluation harnesses required for production use.
Cloud, DevOps & MLOps Execution: Apply cloud-native services, secure deployment patterns, observability, lifecycle management, and MLOps practices to ensure applications are scalable, resilient, cost-aware, and supportable.
Knowledge Transfer & Enablement: Document solutions, mentor receiving teams, and support clean transition of application code, integration patterns, data models, and operational practices into long-term engineering ownership.
Requirements
Bachelor’s or Master’s degree in Computer Science, Software Engineering, or a related quantitative field.
9+ years of hands-on software engineering experience building, deploying, and supporting robust production applications.
Strong proficiency in Python, Java, or Go, with demonstrated ability to apply sound software engineering, testing, data modelling, and system design practices.
Proven experience translating complex business requirements into cloud-optimized application architectures, scalable data models, and technical specifications for production delivery.
Experience operating in a forward-deployed or internal client-facing engineering model, including workflow discovery, stakeholder engagement, rapid prototyping, solution shaping, pilot execution, adoption measurement, and production handoff to receiving teams.
Extensive experience with major cloud platforms such as AWS, Azure, or GCP, including serverless, containerization, managed services, automated deployment, monitoring, and cloud security standards.
Experience designing cloud-based AI solution architectures that integrate managed AI services, LLM providers, retrieval and vector search, secure data access, API gateways, event-driven workflows, identity and entitlement controls, observability, deployment automation, and cost and resilience trade-offs across AWS, Azure, or GCP.
Demonstrated experience integrating LLM or AI/ML capabilities into production applications, including RAG pipelines, embeddings, vector databases or search indexes, prompt templates, context assembly, structured outputs, response validation, evaluation datasets, and model performance monitoring.
Excellent communication and collaboration skills, with the ability to engage technical and non-technical stakeholders, lead cross-functional delivery, and enable receiving teams through documentation, mentoring, and knowledge transfer.
Preferred Experience
Experience building agentic AI systems that decompose tasks, plan multi-step workflows, call approved tools or APIs, maintain state, enforce permission and policy checks, handle failure paths, and produce auditable action trails.
Experience implementing agent observability and controls, including trace capture, prompt and tool-call logging, hallucination and policy-violation checks, human-in-the-loop approvals, incident handoff, and post-action summarization.
Preferred: Experience optimizing production AI systems through model selection and routing, prompt compression, retrieval tuning, caching, batching, streaming responses, asynchronous execution, parallel tool calls, latency budgets, cost controls, and quality regression testing.
ABOUT GOLDMAN SACHS
Answer a McCoy challenge. Companies hiring in engineering see your response and reach out when they want to talk.
About 15 minutes to prepare and 5 minutes to record. Unlimited retakes and a teleprompter help you answer. No AI training, and you can delete your response anytime.
This isn't an application to Goldman Sachs.
goldmansachs:185536_GS_MID_CAREERhttps://api-higher.gs.com/gateway/api/v1/graphql{
"reason": "not_in_rolesearch",
"source_id": "185536"
}See how we measure for definitions, or our corrections log for known issues. Found something wrong? Flag a correction.