Job Description

We're looking for a hands-on Data Scientist or ML Engineer with strong expertise in unsupervised learning, anomaly detection, and causal inference. You'll work on real-world customer support signals--chat, voice, email, social--to uncover patterns, forecast trends, and recommend actions through intelligent, production-ready models.

Key Responsibilities

Pattern Discovery

: Build unsupervised/semi-supervised models (e.g., HDBSCAN, metric learning, spectral) to surface issue clusters.

Trend Detection

: Design anomaly detectors and forecast issue volumes using tools like Prophet or NeuralProphet.

Cross-Channel Correlation

: Link customer signals across support channels to trace issue journeys.

Root Cause Analysis

: Use causal graphs or lightweight GNNs to identify drivers and recommend fixes.

Active Learning

: Develop workflows allowing analysts to refine model clusters and feedback loops.

Experimentation

: Define success metrics, run controlled experiments, and communicate results clearly.

Collaboration

: Work closely with LLM, platform, and dashboard teams to deliver and iterate on features.

Required Qualifications

4+ years in data science/ML roles (bonus for NLP or time-series experience). Strong with clustering, anomaly detection, causal inference. Proficient in Python (pandas, scikit-learn, PyTorch/TensorFlow) and SQL. Experience productionizing models (e.g., via Airflow, Flyte). Familiarity with vector databases and similarity search (e.g., Faiss, PGVector). Experimental mindset with clear communication and incremental delivery focus.
Job Type: Full-time

Pay: ?600,000.00 - ?1,200,000.00 per year

Schedule:

Day shift
Experience:

Data science: 4 years (Required) Machine learning: 4 years (Required)
Work Location: Remote