Artificial Intelligence Engineer - LLM/RAG

Description :

About the Role :

We are looking for an AI Engineer to build and scale data-driven AI systems across Credit Risk, Fraud, Sales, and Collections.

In this role, you will work on transforming raw data into features, embeddings, and knowledge systems that power ML models and GenAI applications (LLM/RAG).

Youll collaborate with data scientists and engineers to build production-ready ML pipelines and intelligent systems.

What Youll Do

Data & Feature Engineering :

- Build and manage data pipelines from multiple sources (transactions, CRM, bureau, etc.)

- Create features & behavioral signals for ML models

- Work with structured & unstructured data

ML Systems & Pipelines :

- Develop scalable ML pipelines (batch + near real-time)

Work on :

1. Feature stores

2. Model registry

3. CI/CD for ML

- Support embedding generation workflows (text, customer, device, etc.)

Experimentation & Model Ops

- Run training & inference jobs

Perform :

- Error analysis

- Model evaluation

- Data quality checks

- Support deployment & monitoring

GenAI / RAG (LLM Enablement) :

- Work on document processing (chunking, cleaning, tagging)

- Support vector search & retrieval systems

- Maintain prompt templates & evaluation datasets

Evaluate :

- Retrieval quality

- Answer quality

Engineering Best Practices :

- Write clean, production-ready code

- Follow Git, CI/CD, and testing practices

- Maintain documentation and ensure data security & compliance

Must-Have Skills :

Programming :

- Strong in Python & SQL

- Good understanding of Git & testing

Data & ML :

- Experience with Pandas / PySpark

Strong in :

- Joins, aggregations

- Feature engineering

ML basics :

- Supervised/unsupervised learning

- Model evaluation

Embeddings :

Engineering Mindset :

- Debugging & problem-solving

- Understanding of ML pipelines & reproducibility

- Logging & monitoring basics

Good to Have :

- ML frameworks : PyTorch / TensorFlow

- MLflow or experiment tracking tools

RAG/LLM stack :

- Vector DBs

- Hybrid search

Tools :

- Airflow / Prefect

- Spark

- Elasticsearch / OpenSearch

- MongoDB

Preferred Experience :

- Experience building end-to-end ML systems

- Exposure to fraud/risk/fintech use cases

- Knowledge of Graph ML / entity resolution

- Experience working on large-scale production systems

Eligibility :

- 5+ years in Data Science / ML Engineering / AI

- Bachelors/Masters in CS / Engineering / Mathematics

Why This Version Works Better

- Simple language - Attracts more candidates

- Clear structure - Easy to scan

- Keywords optimized - Better search results

- Balanced tech depth - Appeals to both ML Engineers & Data Scientists