resume

Education, experience, selected projects, and research.

Divyanshu Pabia

Systems Engineer ยท AI Systems at Scale

Professional Summary

I am focused on understanding and improving the systems that run modern AI models, from GPU kernels and memory behavior to compilers, inference runtimes, and distributed execution. I enjoy reading systems code, profiling bottlenecks, and building experiments that turn performance questions into measurable results. My background spans backend engineering, iOS development, and network-monitoring software for Tor, and I am now specializing in AI systems, GPU computing, and efficient LLM inference.

Experience

  • 2025 - 2026

    Remote Contract

    Network Health Team
    The Tor Project
    • Contributed to reimplementing the Java Tor Metrics Library in Rust to retrieve, parse, and validate network descriptors from CollecTor, focusing on memory safety and extensible parsing.
    • Worked on a unified descriptor pipeline using object storage, a dedicated parser service, and PostgreSQL and VictoriaMetrics exports to support historical and real-time network analysis.
  • 2024 - 2025

    Mysuru, Karnataka

    iOS and backend development intern
    Infosys
    • Built a Hospital Management System with a Swift iOS client and Java/Spring Boot REST APIs for staff management, bookings, and real-time updates. Implemented JWT authentication and role-based access.
    • Containerized backend services with Docker and supported test/staging CI/CD. Strengthened API design, OpenAPI documentation, and verification through JUnit 5 and integration tests.
  • 2023 - 2024

    Chennai, Tamil Nadu

    Apple and Infosys iOS App Development Program
    iOS App Development Centre
    • Built Gamergrid through the Apple and Infosys program, integrating esports events, scores, and news into a Swift iOS app. Developed practical skills in URLSession, Combine, REST API integration, pagination, and resilient data fetching.
  • 2022 - 2022

    Chennai, Tamil Nadu

    Software development intern
    Tech Analogy
    • Built an expense-management backend with Spring Boot and Hibernate, using Kafka and RabbitMQ for asynchronous communication. Gained hands-on experience in transaction processing, microservices, and event-driven design.

Education

  • 2021 - 2025

    Chennai, India

    B.Tech
    SRM Institute of Science and Technology
    Computer Science and Engineering

    CGPA: 8.5/10.

Projects

  • Developed a geospatial assistant prototype combining retrieval over reference material, language-model interaction, and GIS tooling. Explored how to connect natural-language requests with spatial data and tools while keeping analysis steps inspectable.

  • Implemented a virtual last-in, first-out character device as a loadable Linux kernel module. Worked with module and device interfaces, device-state management, and the user/kernel boundary to understand how applications interact with kernel-backed services.

  • Built a VHDL microarchitecture that executes a subset of ARM instructions, connecting instruction decoding, datapath behavior, and control logic. Developed a practical understanding of how instruction semantics map to hardware execution.

  • Built a native Swift app for following esports events, live scores, match statistics, and personalized news. Integrated third-party REST APIs using URLSession and Combine, with attention to pagination, resilient networking, and a responsive mobile interface.

Skills

Programming and systems: C/C++, Python, Java, Rust, Linux, Git, CMake
ML systems: PyTorch, JAX, Transformers, torch.compile, vLLM
GPU and performance: CUDA C++, Triton, parallel computing, profiling, Nsight Systems / Compute
Backend and cloud: Spring Boot, PostgreSQL, REST APIs, AWS, Docker, Kubernetes, Terraform
iOS development: Swift, Xcode, URLSession, Combine
Mathematics: Linear algebra, probability, calculus, optimization