Principal Data Scientist, SIP

About Invesco

As one of the world’s leading independent global investment firms, Invesco is dedicated to rethinking possibilities for our clients. By delivering the combined power of our distinctive investment management capabilities, we provide a wide range of investment strategies and vehicles to our clients around the world. If you're looking for challenging work, intelligent colleagues, and exposure across a global footprint, come explore your potential at Invesco.

Your Role

You Will Be Responsible For:

• Leading and collaborating with software engineers, data engineers, and investment professionals to build innovative AI-based tools using structured and alternate data sources.

• Designing and implementing complex reasoning loops using frameworks like LangGraph, LangChain, or AgentCore, transforming static LLMs into dynamic, goal-oriented agents.

• Developing robust evaluation frameworks and "gold-standard" datasets to measure the performance, faithfulness, and accuracy of generative outputs.

• Scaling AI initiatives and agentic development across Invesco Technology teams by creating reusable patterns, prompt templates, agentic harnesses, skills, and subagents with GitHub Copilot and Claude Code.

• Collaborating with full stack engineers to integrate AI models into business workflows, using AI-assisted coding to move from a notebook experiment to a functional microservice in record time.

• Implementing systematic testing (e.g., RAG evaluation, stress testing) to ensure AI solutions meet the high compliance and reliability standards of the financial services industry.

• Applying emerging techniques in prompt engineering, fine-tuning, and multi-agent systems to solve unique business challenges.

The Experience You Bring:

• 7-10 years of demonstrated experience in data science or machine learning, with a strong emphasis on Natural Language Processing (NLP) and Generative AI.

• Strong foundation in Python and cloud-native development; you write clean, modular code and are comfortable with Docker, CI/CD, and DevSecOps.

• Exposure to or high enthusiasm for learning LangGraph, LangChain, Copilot Studio, or AgentCore with Strands.

• Familiarity with creating evaluation metrics and building datasets for model validation and testing.

• Proactive use of AI tools (GitHub Copilot, Cursor, etc.) to accelerate technical tasks and explore new libraries.

• A desire to work at the intersection of data engineering and full-stack development, with the agility to pivot based on rapid sprint feedback.

Full Time / Part Time

Full time

Worker Type

Employee

Job Exempt (Yes / No)

Yes

Workplace Model

Pursuant to Invesco’s Workplace Policy, employees are expected to comply with the firm’s most current workplace model, which as of October 1, 2025, includes spending at least four full days each week working in an Invesco office. This reflects our belief that spending time together in the office helps us build stronger relationships, collaborate more easily, and support each other’s growth and development.

The above information on this description has been designed to indicate the general nature and level of work performed by employees within this role. It is not designed to contain or be interpreted as a comprehensive inventory of all duties, responsibilities and qualifications required of employees assigned to this job. The job holder may be required to perform other duties as deemed appropriate by their manager from time to time.

Invesco's culture of inclusivity and its commitment to diversity in the workplace are demonstrated through our people practices. We are proud to be an equal opportunity employer. All qualified applicants will receive consideration for employment without regard to race, creed, color, religion, sex, gender, gender identity, sexual orientation, marital status, national origin, citizenship status, disability, age, or veteran status. Our equal opportunity employment efforts comply with all applicable U.S. state and federal laws governing non-discrimination in employment.