Data Scientist - People Analytics (AutoPH Team)

About HighRadius!

HighRadius provides a single Agentic AI platform for the Office of the CFO. It integrates 180+ agents that orchestrate end-to-end processes across Order-to-Cash, Close & Reconciliation, Consolidation & Reporting, Accounts Payable, B2B Payments, and Treasury. HighRadius guarantees operational KPI improvements by mapping them to specific agents on the platform. With a 3-6 month go-live period, HighRadius drives value creation at 1300+ enterprises such as 3M, Unilever, Bristol-Myers Squibb Company, Red Bull, Lufthansa, and more. HighRadius has been consistently recognized as a market leader by Gartner, IDC, and Forrester. 

Job summary

We are redefining how performance is measured and rewarded. As a Data Scientist within the AutoPH Strategy Team , you won’t just be analyzing data—you will be architecting the algo that drives our next-generation Performance Management System. 

Contributing directly to the CEO’s strategic vision, you will engineer the correlation between behavioral inputs and business outputs, and build the algorithms that ensure our organization remains a meritocratic, high-growth environment.

Key responsibilities

1. Algorithm Engineering & PMS Innovation

The AutoPH Engine: Own the statistical logic behind our "Autonomous Performance@ HighRadius" system. You will design, test, and iterate on algorithms that balance Output (results), Input (effort), and Culture KPIs.

Model Evolution: You will critically evaluate our "AutoPH" logic and identify where our measurement systems lack precision. You will  engineer the enhancements required to make our performance algorithms mathematically sound and bridge the gap between what we currently track and what we actually aim to measure. 

Predictive talent strategy: Move from reactive reporting to proactive intervention. You’ll implement ML frameworks (Time Series, Clustering, LLMs among others) to predict talent density, flight risks, and high-potential trajectories.

From Concept to Code: Develop Python scripts to transition theoretical HR models into deployable ML APIs that integrate seamlessly with our business stack.

2. Statistical Rigor & Correlation Research

Cracking the Input/Output Code: - You will apply advanced statistical methods (correlation studies, causal inference, multivariate analysis among others) to prove (or disprove) which "effort" metrics actually move the needle on business outputs.

Bias Mitigation: Use advanced normalization and cohort analysis to eliminate "noise" and systemic bias in performance grading, ensuring a level playing field for every employee.

Data Integrity: Define the taxonomy for performance data. You’ll ensure that our data architecture is clean, scalable, and audit-ready.

3. Productized Analytics & Storytelling

Insight Systems: Don’t just build "reports"—build products. Architect scalable, intuitive Power BI/Tableau dashboards that allow leaders to self-serve insights.

Narrative Building: Translate high-dimensional math into low-friction business narratives for the C-suite. You are the bridge between raw data and strategic execution.

Skill and Experience

The Toolkit: Solid foundation in Python and SQL and experience within the modern ML stack (Scikit-learn, TensorFlow/PyTorch) and BI tools (Power BI/Tableau).

The Experience: 5+ years in Data Science, with at least 2-3 years preferably embedded in People Analytics or Talent Product environments. You understand the nuances of "human data."

The "Product" Mindset: You’ve moved past simple data requests. You have experience identifying "pain points" in a system (like a legacy PMS) and engineering a technical solution to fix them.

Mathematical Grit: You are comfortable with complex variables, cohort normalization, and the challenge of quantifying the "unquantifiable" aspects of human performance.

Why This Role?

High Visibility: This isn't a back-office HR role. You are building a core piece of our company’s operating system under a CEO-led initiative.

Complexity: You’ll be dealing with one of the hardest data problems: predicting and optimizing human behavior at scale.

Autonomy: We value "scientific objectivity." You are expected to challenge the status quo if the data suggests a better way forward.