Mahalakshmy Seetharaman

C++11 and Python developer with over three years of experience in crafting production-grade automotive software. Proficient in interpreting formal specifications, including AUTOSAR, and translating them into robust solutions through unit and integration testing. Expertise in leveraging CI/CD pipelines and Agile methodologies to enhance software development processes.

Skilled in identifying and resolving complex bugs, implementing unit tests with GoogleTest/gmock, and improving software quality through debugging and test automation. Experienced with embedded platforms such as QNX, POSIX, Linux, and QEMU.

In addition to embedded development, bring strong backend and AI expertise with Python, SQL, Flask, and REST APIs, along with growing proficiency in large language models (LLMs). Adept in leveraging CI/CD pipelines and DevOps/testing tools such as Jenkins, Robot Framework, Git, and Bazel to accelerate software delivery.

Recognized for collaboration, a strong willingness to learn, and maintaining a positive, proactive attitude. Excels in clear communication, adapts swiftly to new domains, and thrives in diverse teams that prioritize problem-solving and continuous learning. Eager to contribute to innovative projects that bridge embedded systems and AI.

Master’s Thesis, University of Oulu (ongoing): 

Electrical Capacitance Tomography with Coplanar Plate Sensors

·  Implemented image‑reconstruction algorithms in MATLAB (e.g., Linear Back‑Projection, Tikhonov, Landweber, Newton–Raphson, Levenberg–Marquardt).

·  Built sensitivity and capacitance matrices from measurements; compared convergence, quality, and runtime across phantoms.

LLM Tooling and Multimodal Exploration

·  MCP‑based retrieval for local repositories; LangChain/LangGraph agents for coding assistance and structured Q&A.

·  Experiments toward multimodal pipelines that combine vision and language models for behavior understanding.

Academic Projects (University of Oulu)

·  Design and Implementation of Flight Booking System using RESTful APIs
Developed a group project (3 members) in Python with an SQLite database. Implemented booking workflows via REST APIs, version control using Git, and collaborative development through Bitbucket.

·  Implementation of Zero Mean Normalized Cross Correlation (ZNCC) Algorithm in C with OpenCL Parallelization
Built an initial single-threaded C program for image similarity measurement using ZNCC. Enhanced the project with OpenCL to leverage multiple processors for improved performance. Version-controlled the project using Git on Bitbucket.

Fueling AI with Public Displays? A Feasibility Study of Collecting Biometrically Tagged Consensual Data on a University Campus.

ACM International Symposium on Pervasive Displays (PerDis'19), June 12–14, 2019, Palermo, Italy.

ACM, New York, NY, USA, 7 pages.

My contribution is mainly in the Metadata Analysis part. In order to statistically analyze the quality of the faces collected in the videos and their usability for face biometrics and computer vision in general, Used a state-of-the-art face detector, based on the SSD-framework implemented in OpenCV Python.