Posted 15 days ago
No clicks
**Summary:** The **Software Engineer (Agentic Systems) - Associate** role at Goldman Sachs, Global Investment Research (Dallas) involves building, enhancing, and deploying intelligent research tools and automated agentic pipelines. Responsibilities include: - Developing end-to-end full-stack features using **Java or Python** - Implementing advanced RAG components and integrating AI services using **Google ADK**, **LangChain**, **LangGraph**, **AWS Bedrock**, or similar technologies - Evaluating contextual relevance, groundedness, and hallucination reduction using tools like **Braintrust**, **AgentCore Eval** - Deploying and connecting AI components with enterprise services on **AWS**, configuring **Lambda**, **ECS/Fargate**, **EventBridge**, **S3**, **IAM policies**, and **API Gateway** - Monitoring model and service performance, and triaging production incidents Candidates should have a **Bachelor’s degree in Computer Science, Software Engineering**, or a related quantitative field, along with: - **3–5 years of hands-on software engineering experience** with proficiency in **Java or Python** and modern full-stack web technologies - Practical experience building Gen AI applications using **Google ADK**, **LangGraph**, **AWS Bedrock**, and implementing evaluation pipelines with **Braintrust** and **AgentCore Eval** - Working knowledge of core **AWS services** like **Lambda**, **ECS/Fargate**, **S3**, **EventBridge**, **CloudWatch**, and **API Gateway** Experience with financial domain data, investment research workflows, and token optimization strategies is a plus. The role offers competitive benefits packages.
- Compensation
- Not specified
- City
- Dallas, New York City, London, Birmingham, Singapore, Bengaluru
- Country
- United States, United Kingdom, Singapore, India
Currency: Not specified
Full Job Description
What We Do
At Goldman Sachs, our Engineers don't just make things we make things possible. Change the world by connecting people and capital with ideas. Solve the most challenging and pressing engineering problems for our clients. Join our engineering teams that build massively scalable software and systems, architect low latency infrastructure solutions, proactively guard against cyber threats, and leverage machine learning alongside financial engineering to continuously turn data into action. Drive new businesses, redefine finance using AI, and seize opportunities at market speed.
In an era defined by AI, Engineering isn't just central it's the driving force behind our business. Our dynamic environment demands innovative strategic thinking and immediate, impactful solutions. Ready to push the limits of digital possibility? Begin your journey here.
Who We Look For
Goldman Sachs Global Investment Research (GIR) is seeking an Agentic Systems Engineer to build, enhance, and productionize intelligent research tools and automated agentic pipelines. You will collaborate with senior engineers to develop end-to-end full-stack applications, implement advanced RAG components, and integrate AI services into mission-critical GIR platforms.
How You Will Fulfill Your Potential
- Feature Engineering: Build and maintain end-to-end full-stack features, integrating analyst-facing frontends with backend APIs and microservices implemented in Java or Python.
- Gen AI & Agent Implementation: Implement and optimize advanced RAG pipelines and stateful multi-agent workflows using Google ADK, LangChain, LangGraph, AWS Bedrock, or equivalent technologies, incorporating structured outputs, tool calling, and human-in-the-loop steps.
- Evaluation & Benchmarking: Implement automated offline and online evaluation suites using tools such as Braintrust, AgentCore Eval, equivalent industry solutions, or custom test harnesses to assess contextual relevance, groundedness, and hallucination reduction.
- Cloud & Data Integration: Deploy and connect AI components with enterprise services on AWS, configuring Lambda, ECS/Fargate, EventBridge, S3, IAM policies, and API Gateway endpoints.
- Observability & Incident Support: Monitor model and service performance using CloudWatch and distributed tracing; triage production incidents, conduct root-cause analyses, and optimize token/resource consumption.
- SDLC Discipline: Actively participate in peer code reviews, write automated unit and integration tests, maintain CI/CD pipelines, and document integration patterns and data models.
- Agile Collaboration: Coordinate with business stakeholders and cross-functional teams in Agile sprints to translate research requirements into deliverable technical stories.
Basic Qualifications
- Bachelors degree in Computer Science, Software Engineering, Applied Mathematics, or a related quantitative field.
- 35 years of hands-on software engineering experience with proficiency in Java or Python, combined with modern full-stack web technologies (e.g., React, TypeScript, FastAPI, Spring Boot).
- Practical experience building Gen AI applications using Google ADK, LangGraph, or equivalent agentic frameworks, vector databases, and AWS Bedrock.
- Experience implementing evaluation pipelines and prompt testing with Braintrust, AgentCore Eval, equivalent technologies, or custom benchmark harnesses.
- Working knowledge of core AWS services, including Lambda, ECS/Fargate, S3, EventBridge, CloudWatch, and API Gateway.
- Solid understanding of standard SDLC tooling, automated testing, Git, Docker, and CI/CD automation.
Preferred Qualifications
- Exposure to financial domain data, investment research workflows, or enterprise content authoring tools.
- Practical understanding of token optimization, prompt caching, and latency reduction strategies.
- Familiarity with guardrail frameworks defending against prompt injection and unauthorized data access.
- Experience building asynchronous or event-driven microservices across polyglot Java and Python stacks.
Please note we will only respond to applications for which we have an interest.
Benefits
Goldman Sachs is committed to providing our people with valuable and competitive benefits and wellness offerings, as it is a core part of providing a strong overall employee experience. A summary of these offerings, which are generally available to active, non-temporary, full-time and part-time US employees who work at least 20 hours per week, can be found here.
