Resume

Pranay Meshram

pranaywa@buffalo.edu | linkedin.com/in/pranayspeed | github.com/pranayspeed | pranayspeed.com

EDUCATION

  • Master of Science: Engineering Science Robotics, University at Buffalo, The State University of New York, Expected December 2022

  • Bachelor of Engineering: Computer Science Engineering, Rashtrasant Tukadoji Maharaj Nagpur University, December 2014

WORK EXPERIENCE

Research Assistant, University at Buffalo, NY

(May 2022 - Present)

  • Modular and Extensible Visual SLAM framework: Rust
    • Design and implement a modular Visual SLAM framework based on actor model, to help researchers in advanced research in Visual SLAM without re-inventing wheel
    • Conducting independent research in Drones Lab, focused on different Structure from motion and Visual Odometry algorithm
  • UAV Last Mile Delivery: Python, FPGA, GPU
    • Research towards Low Power Object detection design and Deployment in FPGA and GPU
    • Perform study for various Convolution Neural Network design and perform research in optimizing power consumption, to deploy it in UAV for real time object detection
    • Research and devise safety aware and robust path planning algorithm for UAV

System Software Engineer, NVIDIA Graphics Pvt . Ltd, Pune, India

(December 2018 - May 2021)

  • Developed solutions and tested for Nvidia GPU Compiler Verification reducing engineering effort by 20 percent.
  • Led development of concept to design and utilize Compiler to verify Compiler for Automotive Safety Project to satisfy ISO 26262
  • Led development of CUDA language program generating Fuzzer using Deep learning
  • Experienced in working towards Automotive Safety certification for CUDA Assembler (ISO 26262)

Software Engineer, Renishaw Metrology Systems Ltd, Pune, India

(July 2017 - December 2018)

  • Researched and created solution for automated GD&T and contribute towards comprehensive Metrology SDK helpful for automatic manufactured part inspection
  • Identified and fixed major and legacy product bugs pending for years, by introducing a combination of optimization techniques, eventually resulted in zero bugs.
  • Conducted research in optimization algorithms including Gradient Descent, Levenberg Marquardt, Simplex Method, DownHill Simplex Method, simulated Annealing and Particle Swarm optimization select one of best combination
  • Researched and devised Proof of Concept on Voxel based Motion path planning algorithm to minimize robot arm motion in vertical direction to almost zero
  • Handled Probe compensation and Extraction algorithms, Statistical Analysis, Vector algebra, Geometric fitting and tolerancing problems

Software Developer, ProtoTech Solutions & Services Pvt. Ltd, Pune, India

(February 2015 - June 2017)

  • Awarded with Rising Star of the Year 2015 - 2016 for best performance as a new graduate
  • Researched and implemented in-house polygon reduction feature, saved licensing cost of leveraging 3rd party commercial library, as well as helped close a deal for around 50+ licenses of product with a well-known industrial robotics company
  • Created Generic Assembly structure to cope up with translation from one assembly hierarchy to another also eased features implementation, now deals with single generic assembly rather than earlier 3 different hierarchies
  • Maintained and enhanced a Major Integrated CAD formats translator Application and drove major code refactor and revamping old functional modules
  • Steered Architecture Designing of C++ plugins for AutoCAD, Revit and Inventor
  • Built automated mechanism for creation of installer scripts and installers, by the Plugin directory name and all 40+ plugins are now built with single click to script
  • Planned and designed complete Architecture for a generic Plugin Factory project, leveraging existing tools and Achieved code redundancy next to negligible and was able to support 40+ plugins

RESEARCH EXPERIENCE

Journal Publications

  • Pranay Meshram, Mohanish Indurkar, Ritu Raj, Neeranjan Chitare
    Automated License Plate Recognition System (ALPRS) using Regular Expression International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 International Conference on Industrial Automation And Computing (ICIAC- 12th & 13th April 2014)[PDF]

Posters

  • Sofiya Semenova, Pranay Meshram , Timothy Chase, Steven Ko, Yu Liu, Lukasz Ziarek, Karthik Dantu. (2022). A modular, extensible framework for modern visual SLAM systems. Conference: MobiSys ‘22: The 20th Annual International Conference on Mobile Systems, Applications and Services 579-580. 10.1145/3498361.3538793.[Link]

TECHNICAL SKILLS

Programming : C, C++, C#, Java, Python, Rust
Technical: Matlab Experience, ROS(Robot Operating System), Tensorflow, Pytorch, Visual SLAM, Computer Vision, Robot Kinetics and Kinematics, Machine Learning
Database : MS SQL Server

ACHIEVEMENTS

  • DAC System Design Contest 2022 (June 2022) - 4th place [UBPercept] Results

    Contirbution

    • Led the team of 7 and contributed towards initial investigation and implementation.
    • Worked in a team to create plan for CNN training and deployment to FPGA (Ultra96V2)
    • Major contribution in automating the deployment and result analysis for fast and efficient experiments.