Lead Data Scientist – Algorithmic Defense

Lewis & Grey International; Sydney; Hybrid; AUD 180,000 - 320,000; Full time

Benefits offered by this job

Equity participation

Competitive compensation package

Job summary

Lewis & Grey International seeks a Lead Data Scientist for Algorithmic Defense to reverse engineer black box algorithms and build defensive infrastructure protecting clients from bias and synthetic fraud. You will lead the development of our proprietary Forensic Content Audit tools and Knowledge Graph strategies to drive defensible, data-driven outcomes.

You will work across NLP, ML theory, and graph databases to quantify reputational alpha and defend clients in complex regulatory environments.

Qualifications

• PhD or Masters in CS/Statistics/Mathematics/Computational Linguistics.
• Strong ML theory and NLP background beneficial.
• Experience with adversarial thinking and reputation systems preferred.

Responsibilities

• Adversarial Algorithm Analysis: study ranking factors and moderation policies; model calcification velocity of negative content.
• Develop and deploy Knowledge Graph Injection strategies to counter AI hallucinations.
• Synthetic Identity & Bot Detection: build classifiers to distinguish real vs synthetic reviews.
• Quantifying Reputational Alpha: correlate sentiment with financial metrics for ROI analysis.
• Lead the development of the Forensic Content Audit tools and the Lewis & Grey Intelligence Platform.

Skills

Adversarial thinking

Mathematical ML

Economic insight

Linguistic precision

Python

R

SQL

Education

PhD in CS/Statistics/Math/Computational Linguistics

Tools

Python

R

SQL

Hugging Face

spaCy

Pinecone

Milvus

Neo4j

Job description

Lewis & Grey is a strategic advisory firm operating at the intersection of technology and law and behavioral finance. We are not a digital marketing agency. We are a reputation defense firm that treats public perception as a financial asset. We protect venture capital firms and medical practices and sovereign entities from the volatility of the digital ecosystem. Our methodology is built on the premise that online reputation is an engineering problem rather than a creative one. We use data precision to challenge unfair reviews and displace AI hallucinations and restore the congruency between operational reality and public narrative.

The Mandate

The Lead Data Scientist for Algorithmic Defense is the technical anchor of our intelligence division. You will not be optimizing ad spend or building recommendation engines. You will be reverse engineering the black box algorithms that govern corporate reputation. Your mandate is to build the defensive infrastructure that protects our clients from algorithmic bias and synthetic fraud and coordinated attacks.

In this role you will lead the development of our proprietary "Forensic Content Audit" tools. You will design the systems that scan the internet for policy violations and social signals. You will turn the abstract concept of reputation into a quantifiable metric that can be defended in a boardroom and utilized in legal appeals.

Primary Responsibilities

Adversarial Algorithm Analysis

You will continually analyze the ranking factors and moderation policies of major search engines and review platforms. You will deploy reverse engineering strategies to understand how negative signals calcify into permanent records. You will build models that predict the "calcification velocity" of negative content allowing our analysts to intervene before a signal creates permanent damage.

You will lead the firm's technical response to generative AI risk. You will build vector based monitoring systems that detect when Large Language Models begin to hallucinate false narratives about our clients. You will design the "Knowledge Graph Injection" strategies that flood the ecosystem with structured high authority data to force the models to recalculate their probabilistic outputs.

Synthetic Identity & Bot Detection

You will engineer the immune system against the "Digital Identity Collapse." You will develop classifiers capable of distinguishing between genuine human feedback and synthetic reviews generated by bot farms. You will analyze linguistic patterns and metadata anomalies to flag coordinated smear campaigns that leverage AI generated content.

Quantifying Reputational Alpha

You will work with our financial advisory practice to build the "Reputation Valuation Model." You will correlate sentiment data with financial performance metrics such as Customer Acquisition Cost and stock price volatility. You will create the mathematical proof that demonstrates the ROI of reputation defense to a CFO.

The Candidate Profile

We require a practitioner who views data science as a form of counter intelligence. The ideal candidate is an adversarial thinker who understands both the mathematics of machine learning and the economics of information.

Academic Rigor

A PhD or Masters in Computer Science, Statistics, Mathematics, or Computational Linguistics. You are comfortable with the theoretical underpinnings of high dimensional vector spaces and probabilistic modeling.

Technical Stack

Mastery of Python and R and SQL. Deep experience with NLP libraries (Hugging Face, spaCy) and vector databases (Pinecone, Milvus). Experience with graph databases (Neo4j) is highly preferred.

You possess a deep understanding of Transformer architectures and retrieval augmented generation (RAG). You understand how training data contamination works and how to counter it.

Forensic Experience

Experience in fraud detection, cyber security, or trust and safety at a major platform is a distinct advantage. You understand how bad actors manipulate systems.

Core Competencies

Zero Trust Logic

You do not accept data at face value. You assume every signal is potentially synthetic until proven otherwise. You design systems that operate effectively in a high noise and low trust environment.

Linguistic Precision

You understand that language is data. You can parse complex Terms of Service into logical rules that can be automated. You can translate legal concepts into code.

Commercial Translation

You can explain complex algorithmic threats to non technical stakeholders. You can look at a confusion matrix and explain what it means for the client's bottom line.

Ethical Fortitude

You will be operating on the front lines of the information war. You must adhere to the highest standards of data privacy and ethical conduct. We protect the truth. We do not manufacture it.

The Value Proposition

This is not a maintenance role. It is a build role.

You will build the "Lewis & Grey Intelligence Platform" from the ground up. You will choose the stack and define the roadmap.

You will be solving problems that do not yet have Stack Overflow answers. You will be defining the defense mechanisms for the post generative era.

High Impact

Your code will directly protect the livelihoods and legacies of high performing individuals. You will see the immediate result of your work in the restoration of trust.

Location & Compensation

The role is based in our primary technical hub. We offer a compensation package designed to compete with top tier technology firms including significant equity participation. We reward the ability to solve the unsolvable.