Available for Opportunities

Sindhu
Satish Kumar

Software & ML/Data Engineer. I build production-grade pipelines, ML inference systems, and scalable data infrastructure, with a focus on reliability, performance, and real-world impact.

Get in touch LinkedIn GitHub

Who I am

I'm a recent MSCS graduate from USC with 3+ years of experience building data-intensive systems, ML pipelines, and scalable software. I care deeply about building things that work in the real world. Not just in notebooks.

My background spans full-stack data engineering, deep learning research, and production software development. I've worked across distributed systems, CI/CD automation, and ML model development, always with an eye on reliability, observability, and performance.

I'm currently seeking SDE, ML Engineer, and Data roles where I can contribute to meaningful products and continue growing as an engineer.

3+years of experience
90%code coverage achieved
0.90Dice score on medical imaging
40%build time reduction

Where I've worked

May 2024 — Present
ML Researcher
University of Southern California, Los Angeles
  • Designed a scalable deep learning inference pipeline using ResNet-18 to process 2,200+ dermatological images, covering the full workflow from preprocessing through model evaluation
  • Validated and optimized model accuracy using hybrid loss functions (BCEWithLogits + Dice), achieving a 0.90 Dice score and improving reliability by 12% over baseline
  • Engineered data augmentation workflows to improve DNN inference robustness across real-world inputs
Feb 2021 — June 2023
Software Development Engineer
Sixt, India
  • Decomposed a monolithic frontend into Micro-Frontend services, enabling independent deployments that reduced build times by 40%
  • Built a telemetry and observability system to capture high-velocity user interaction data, improving search ranking algorithms and increasing engagement by 15%
  • Resolved memory and rendering bottlenecks, increasing execution speed by 16% and raising test coverage from 65% to 90% via CI/CD
  • Mentored junior engineers and standardized onboarding docs, reducing ramp-up time by 25%

Academic background

Master of Science in Computer Science
University of Southern California
Aug 2023 — May 2025
Machine Learning · Foundations of AI · NLP · Deep Learning · Information Retrieval · Database Systems
Bachelor of Computer Science
JSS Science and Technology University
Aug 2017 — May 2021
Data Structures · Algorithms · OOP · Operating Systems · Computer Architecture · Data Mining · Statistics

Things I've built

Scalable Data Pipeline — Financial Market Insights
Fault-tolerant distributed pipeline using Airflow and PySpark with automated monitoring, diagnostics, and self-healing workflows.
↓ MTTR by 20% · ↓ config overhead 40%
AirflowPySparkPostgreSQLDocker
3D CNN — Neuroimaging Predictive Modeling
End-to-end 3D CNN inference pipeline using ResNet and DenseNet with custom convolution blocks for volumetric MRI data.
0.99 AUC · MSE improved 8.63 → 8.34
PyTorchResNetTransfer Learning3D CNN
Supply Chain Optimization Pipeline
Automated cloud-integrated pipeline using GitHub Actions and BigQuery for safety stock optimization and shipment delay prediction.
95%+ service level maintained
BigQueryGitHub ActionsPythonZ-score
Transformer-Based Legal Text Summarization
Parallelized inference pipelines for Transformer models (T5, DistilBART) with domain adaptation strategies for legal documents.
↑ ROUGE scores by 15%
T5DistilBARTHuggingFaceNLP

What I work with

Languages
PythonSQLJavaScriptTypeScriptC++Shell
ML / AI
PyTorchTensorFlowHuggingFaceScikit-learn
Data & Cloud
AirflowPySparkBigQueryAWSDockerPostgreSQL
Core Skills
ML PipelinesETL/ELTCI/CDAgileTDDREST APIs

Let's connect

I'm actively looking for SDE, ML Engineer, and Data roles. If you have an opportunity or just want to chat, reach out — I'd love to connect.

✉ sindhusatishkum@gmail.com LinkedIn GitHub