Skip to main content
Suyash Singh

Suyash Singh

Fullstack Engineer with 3 YoE building High Scale Systems

Talent
BengaluruMember since October 19, 2023
PythonC++GoJavaScriptDjangoDjango Rest FrameworkCeleryRedisNestJSDockerAWSAzurePostgreSQLNginxReactFlutterAnsibleKafkaBERTLLM Inference
·

Hire this person

Send a job offer directly to this candidate

PythonC++GoJavaScriptDjango

Experience

Software Engineer - Email Like Messaging Security (ELMS) at Abnormal Security (2025-04 – Present)

Software Engineer working on Email Like Messaging Security

  • Led and shipped the project, attachment malware scanning for Microsoft Teams, protecting 150+ enterprise customers; designed the service to handle 1,500 attachments/hr
  • Scaled ELMS microservices (Python and Golang) 10× (100K → 1M+ messages/hr) by improving service concurrency, introducing multithreading in Python services, caching hot paths with Redis, and optimizing Kafka consumer polling
  • Designed and shipped Behavioral Security for ELMS, built and fine-tuned BERT-based models to detect message intent and malicious patterns, and integrated them into the real-time detection pipeline
  • Led redesign of ELMS frontend pages written in React, improving workflows for messaging security

Software Engineer at Commvault (2024-01 – 2025-04)

Software Engineer working on backend services and internal tools

  • Implemented the retrieval stack for the Internal RAG Tool - Arlie using Azure AI Vector Search and MiniLM-based re-ranking to boost retrieval precision
  • Developed backend services for the HyperscaleX platform using C++ and Python, and contributed to internal tools built with React
  • Contributed to the 1-Touch Linux product including a UI migration from YAST to Electron, modernizing the Linux desktop recovery interface

Open Source Contributor at Code for GovTech (2023-07 – 2023-08)

Open Source Contributor focused on backend performance optimization

  • Drove backend performance optimization across critical APIs, reducing response time from 3.4s to 200ms by eliminating N+1 query patterns, removing redundant SQL joins, and redesigning serializer-driven ORM access
  • Built performance profiling and query analysis workflows using django-silk and custom middleware, enabling faster root-cause analysis of slow endpoints and more systematic performance tuning
  • Improved backend reliability and maintainability by increasing test coverage from 45% to 90% and strengthening API validation with structured test suites and enum-based checks

Education

B.Tech in Computer Science and Engineering – Dr. A. P. J. Abdul Kalam Technical University (2020-12 – 2024-06)

Reviews

Related