Data Scientist (Financial Crimes)

Job Description:

Data Scientist

Financial Crimes Models & Analytics          

Note: Fidelity will not provide immigration sponsorship for this position.

Description

Join Fidelity’s Financial Crimes Models & Analytics team, where we use innovative data science techniques to protect the firm and its clients from financial crime. The Financial Crimes Models & Analytics team, within Fidelity’s Enterprise Compliance group, is responsible for the design, development and optimization of the firm’s AML detection and sanctions models. We partner with colleagues across Fidelity Enterprise Risk Management to build AI-powered software and data solutions that support Fidelity’s Financial Crimes Compliance programs.

We’re seeking a passionate data scientist to assist with the ongoing transformation of the company’s Financial Crimes Compliance program!

The Team

Our team has broad responsibility for Fidelity Investments transaction monitoring and sanctions systems across multiple business units. Coverage areas include: Digital Asset Trading and monitoring of On-Chain Transfers, Insider Trading, Terrorist Financing, International Money Movement, Global Sanctions & Anti-Corruption, Elder Financial Exploitation, Negative Media screening, Market Manipulation, Securities Fraud, Low-Priced Securities, etc. Our team uses machine learning principles and new technologies while constantly aspiring to improve Fidelity’s surveillance and screening capabilities.

The Purpose of Your Role

As a Data Scientist, you will research, develop, and deliver next generation surveillance solutions on a wide range of AML typologies! These include but are not limited to, digital asset market manipulation, money laundering and crypto scams aka 'Pig Butchering', traditional markets manipulation (equities, options, cryptocurrencies), securities fraud, insider trading, elder financial exploitation, anti-money laundering and terrorist financing.

The Responsibilities of this role

• Design and tune both machine learning and rules-based solutions for the Fidelity Digital Assets business

• Research and develop models that identify suspicious transactions and customers

• Contribute to implementation of LLM-powered solutions in support of the greater Financial Crimes Compliance organization

• Work on multiple long/medium-term data science projects concurrently under moderate direction

• Participate in code reviews to enable learning, collaboration and mentoring of other team members

• Make presentations to update team on project progress, research and new findings

• Collaborate with members of the team as well as external teams on the planning, research, development and productizing of data science solutions

• Stay current with advances in ML/AI, especially in the areas of cryptocurrency, generative AI and financial crime detection

• Document research findings and project progress

The Expertise You Have

• Bachelor’s in Computer Science, Mathematics, Computational Statistics or related field and several years of related experience or a Master’s degree in a related field

• Strong programming skills including 2+ years’ experience with Python and SQL

• Experience carrying out various aspects of a data science project including exploratory analysis, data cleaning, preparation and annotation, ML pipeline design and development, model evaluation and validation

• Experience with LLM frameworks and tools (e.g., LangChain, deepeval)

• Familiarity with RAG architectures, prompt engineering, and fine-tuning techniques

• Experience with libraries such as Spacy, NLTK, Stanford NER, scikit-learn, pandas, tensorflow, keras, pytorch, numpy

• Experience with big data tools such as Spark or snowpark

• Experience working with smaller data sets and a lack of labeled data

• Familiarity with digital assets

• Proven experience with both supervised and unsupervised machine learning algorithms such as decision trees, isolation forests, autoencoders/neural networks, linear/logistic regression, clustering, etc

• Experience with general software tools/frameworks such as git, pytest, dbt

• Experience with most of the following: exploratory data analysis, preprocessing and normalization of data, text wrangling, dimensionality reduction, anomaly detection, rare event modeling, statistical analysis, big data manipulation, language modeling, word embeddings, machine learning pipeline architecture

The Skills You Bring

• A balance of res