Location : Bangalore (Hybrid)
Experience : 4–7 years
Team : Authify – Intelligence & ML
Type : Full-time, Hands-on IC Role
About Authify
At Authify, we build end-to-end intelligence systems that transform raw signals — network data, text, audio, behavioral patterns — into actionable insights. Our ML stack blends graph intelligence, behavioral analytics, offline learning systems, and scalable pipelines to support high-impact, real-time decision workflows.
Role Overview
We are hiring a Senior Machine Learning Engineer with a strong background in anomaly detection, behavioral clustering, threat scoring, graph intelligence, and transcription pipelines . This role focuses on designing ML systems that operate reliably at scale, across billions of events, diverse signals, and heterogeneous data sources. The ideal candidate has built production ML systems, not just academic models.
What You’ll Work On
- Build scalable ML models for graph-based anomaly detection , behavioral pattern analysis , and threat scoring .
- Develop clustering pipelines to identify target groups, detect suspicious patterns, and enrich intelligence datasets.
- Design offline learning pipelines using PyTorch , managing large training datasets and model refresh cycles.
- Build robust Whisper-based transcription pipelines for structured intelligence extraction.
- Work with distributed systems (Kafka → Spark → storage layers) to build ML-ready datasets.
- Implement model evaluation frameworks, drift detection, and continuous improvement workflows.
- Contribute to internal ML libraries, feature stores, and graph intelligence frameworks.
Required Experience
4–7 years of experience in ML engineering, data science, or applied AI in production systems.Strong expertise in PyTorch and ML fundamentals (supervised, unsupervised, clustering, embedding techniques).Experience with graph algorithms (GraphSAGE, GNNs, community detection, link prediction).Strong grounding in offline learning systems , batch retraining, and pipeline orchestration.Experience working with large datasets, distributed compute, and scalable inference systems.Experience building ML for anomaly detection, fraud detection, behavioral modelling, or intelligence systems.Nice to Have
Experience with Whisper or other speech-to-text models.Experience in SOC / telecom / fraud / intelligence analytics (not mandatory).Exposure to Elasticsearch, Neo4j, TimescaleDB, Spark, or Kafka.What We Value
End-to-end ownership from data → model → metrics → deploymentCritical thinking, especially in noisy real-world scenariosHands-on engineering and production mindsetCuriosity to explore new modelling approaches and graph techniques