Machine Learning Engineer – Credit Scoring & MLOps

Overview

About Arbix AI
Arbix AI Solutions is an early-stage deep-tech startup based at T-Hub, Hyderabad, building responsible AI products for India’s agricultural credit ecosystem.
We are a team of seasoned bankers, technologists and AI specialists working to deliver secure, affordable and impactful digital solutions for inclusive and future-ready banking across India.
Our flagship platform, SaakhSetu, helps financial institutions assess rural and agricultural borrowers faster, more transparently and more fairly. At the core of SaakhSetu is Yogyank, our ML-powered scoring engine that generates a bank-agnostic Entitlement Score for underserved rural and agri borrowers.
Our approach combines deep banking domain experience with Explainable AI to build scoring systems that are practical, auditable, and suited to the realities of India’s agricultural credit system.

About the Role

• We are looking for a Machine Learning Engineer – Credit Scoring & MLOps to help build, validate and productionize the Yogyank scoring engine.
• This is not a notebook-only data science role. We need someone who can build reliable ML systems where data pipelines, feature workflows, model artifacts, explainability outputs, and versioning are controlled, documented, and reproducible.
• You will work directly with the founders and engineering team to convert structured and alternative data sources relevant to rural credit assessment into a robust scoring system for financial inclusion.

Key Responsibilities

• Build and maintain end-to-end ML training and scoring pipelines.
• Engineer meaningful features from domain-specific borrower and contextual data.
• Design leakage-safe ML workflows using appropriate validation methods.
• Train and compare suitable tabular ML models.
• Package models, preprocessing logic, schemas, and metadata for reproducible scoring.
• Build explainability outputs that are stable, understandable, and useful for business review.
• Support model monitoring for drift, stability, missingness, segment-level behaviour, and fairness considerations.
• Work with backend engineers to integrate scoring into APIs or batch workflows.
• Document model assumptions, limitations, and validation results clearly.
• Explain ML risks, tradeoffs, and implementation choices clearly to founders, engineers, and business stakeholders.

Required Skills

• 3+ years of hands-on experience in ML engineering, applied ML, data science, or MLOps.
• Strong Python skills.
• Strong experience with pandas, NumPy, and scikit-learn.
• Practical experience with XGBoost, LightGBM, CatBoost, or similar tabular ML models.
• Good understanding of feature engineering, validation design, data leakage, model evaluation, and reproducibility.
• Ability to write clean, modular, reviewable ML code.
• Ability to build production-oriented ML workflows, not just offline experiments.

Strongly Preferred

• Experience in credit scoring, lending, fintech, banking, risk models, or regulated ML.
• Experience with MLOps tools such as MLflow, DVC, model registries, or artifact tracking.
• Experience building production scoring pipelines.
• Experience with out-of-time validation, encoding strategies, model explainability, drift monitoring, or model governance documentation.

Good to Have

• Experience with tabular ML, geospatial ML, alternative-data problems, or domain-specific scoring systems.
• Prior work in agri-tech, rural finance, NBFCs, banks, KCC, cooperative banking, or agricultural lending.
• Experience deploying ML models behind APIs.
• Familiarity with FastAPI, Docker, cloud platforms, CI/CD, or data pipelines.

What We Offer

• Competitive startup compensation.
• ESOP / equity opportunity for the right foundational team member.
• Direct work with founders and domain experts.
• Ownership of the ML core of a high-impact financial inclusion product.
• Opportunity to build responsible AI for underserved rural and agricultural borrowers in India.
• Work from our office at T-Hub, Hyderabad, within one of India’s leading startup ecosystems.

How to Apply
Send your resume, GitHub / portfolio link, and a short note about one ML system, scoring model, or data pipeline you built from scratch to: hr@arbixai.com

In your note, please mention:

• what you built
• what data you used
• how you validated it
• how you prevented leakage
• how you made it reproducible or production-ready