Senior Software Engineer ML Pipelines & Observability

Aldi Fahrezi

SENIOR SOFTWARE ENGINEER

425-***-**** ***********.****@*****.*** Seattle, WA https://www.linkedin.com/in/aldifahrezi PROFESSIONAL SUMMARY

Senior Software Engineer with 6+ years of experience building production data pipelines and ML infrastructure. Strong in Python, distributed systems, and streaming data. Experienced in reducing data latency, improving data reliability, and enabling robust model monitoring. Proven track record of supporting ML workflows from data ingestion to serving with clear ownership and measurable impact. WORK EXPERIENCE

Senior Software Engineer Google Seattle, WA Mar 2021 – Present

• Built batch and streaming feature pipelines for flood forecasting models using Python and Apache Beam, processing over 5 TB of satellite and sensor data daily with 99.9% pipeline reliability

• Designed and implemented core components of telemetry pipelines using Pub/Sub and BigQuery within systems processing over 2B events per day, enabling model performance monitoring and data drift detection

• Reduced data-to-model latency from 6 hours to under 2 hours by productionizing ML outputs and optimizing feature computation workflows

• Implemented model monitoring dashboards and alerting pipelines, reducing issue detection time by 40% and improving response time during incidents

• Developed distributed feature query services in Go and Java, reducing average latency by 25% for online inference use cases

• Improved observability of ML pipelines by increasing telemetry coverage by 30%, enabling faster root cause analysis

• Mentored 4 engineers on ML pipeline design, including idempotent processing, backfills, and validation strategies

Software Engineer Google Singapore Apr 2019 – Mar 2021

• Built event-driven data pipelines using Pub/Sub and Kafka, contributing to systems processing over 200M transactions per day to support fraud detection models

• Reduced end-to-end pipeline latency by 35% through improved batching, partitioning, and parallel processing

• Optimized BigQuery and Bigtable queries, reducing read latency by 20% for feature retrieval and analytics workloads

• Implemented idempotent processing and retry mechanisms, reducing duplicate transaction errors by 60% and improving training data quality

• Defined data contracts and APIs with product and ML teams to ensure consistent feature definitions across training and serving systems

Head of Data Science SIG RISTEK, Universitas Indonesia Jakarta, Indonesia Jan 2018 – Dec 2018

• Led a team of 20 students building ML-based data visualization tools for academic projects

• Delivered workshops on Python, data pipelines, and machine learning fundamentals to over 100 students

• Mentored teams on end-to-end ML workflows, including data collection, feature extraction, model training, and visualization

TECHNICAL SKILLS

Languages: Python, Go, Java, SQL

Machine Learning and Data: Feature engineering, batch and streaming pipelines, model monitoring, data drift detection, training and serving consistency

Frameworks and Tools: Apache Beam, BigQuery, Pub/Sub, Kafka, Dataflow Backend and Systems: Distributed systems, microservices, REST APIs, gRPC, idempotent processing Cloud and Infrastructure: Google Cloud Platform, Docker, Kubernetes, CI/CD Data & Streaming: Apache Beam, Dataflow, BigQuery, streaming and batch processing, ETL pipelines EDUCATION

BACHELOR’S DEGREE Computer Science University of Indonesia 2015 – 2019

Contact this candidate