Embedded Fullstack Developer

My main responsibility was to build the C++ runtime interfaces for a new high performance runtime. I also worked on a project in embedded software, predicting non-functional properties of System-on-Chip Architectures. On a daily basis, I have been using C++ and modern C++ 2011+ features like template meta-programming. In this process, I have played an important role among the developers leading and helping them deliver their tasks. I have also been in close touch with other project partners, stakeholders and managers to ensure good communication and solution adoption.

Organising team meetings outside office hours to strengthen bounds between members

Discuss personal achievements and provide advice and support during lunch time

Champion the agile process followed within the Dev Team. Ensure that our processes are understood by others

Help and support members choosing and maintaining the path towards improving current aptitudes and gaining new skill sets

Worked daily in Linux-based HPC environments, compiling, deploying, and debugging C++ runtimes and parallel applications on multi-node systems.

Integrated the HPX high-performance C++ runtime with heterogeneous computing frameworks such as CUDA, OpenMP, and MPI for large-scale workloads.

Designed and reviewed GSoC project proposals focused on parallel computing, task-based execution models, and large-scale scalability on supercomputers.

Implemented and tuned asynchronous and distributed execution models suitable for massively parallel HPC applications.

Performed in-depth performance analysis and optimization of parallel C++ code, focusing on thread scheduling, load balancing, and memory usage.

Used profiling and debugging tools (e.g. Valgrind, Perf) to identify performance bottlenecks, memory leaks, and concurrency issues in HPC applications.

Designed and executed scalability, stress, and regression test plans to validate runtime stability on high core counts and distributed-memory systems.

Collaborated with HPC users and research teams to reproduce and resolve runtime incidents on multi-node clusters.

Handled service request style support from researchers for HPX and HPC runtime components.

Authored technical documentation and deployment guides for installing, configuring, and operating HPC runtimes on Linux clusters.

Worked with modern C++ (C++11/14/17/20) and template metaprogramming to build efficient, low-overhead abstractions for high-performance and parallel code.

Contributed to improving HPC runtime APIs for better usability, robustness, and efficiency in large-scale scientific and engineering applications.

Collaborated with hardware and computer-architecture researchers to analyse and predict non-functional properties of SoC and HPC-style systems.

Used Python and scientific libraries (NumPy, SciPy, Pandas, scikit-learn) for data analysis, performance evaluation, and result post-processing in HPC research projects.

Used CUDA and OpenMP, MPI to implement Algorithms and Applications on Cluster Nodes

Applied structured incident and problem management practices aligned with ITIL-style process thinking in HPC contexts.

Worked closely with cross-functional teams to ensure reliable operation of HPC software stacks.

Supported contributors and users in debugging distributed and parallel workloads, including race conditions, scalability issues, and non-deterministic failures.

Promoted CI/CD and automated testing for high-performance runtimes to maintain reliability and reduce regressions on HPC systems.