Vineeth Chinmaya Murthy

Vineeth Chinmaya Murthy

Bengaluru, Karnataka, India | vineethcmps@gmail.com

Summary

Highly motivated and experienced AI/ML professional with a proven track record of success in developing and implementing cutting-edge AI solutions. Expertise in various AI/ML frameworks and tools, including TensorFlow, PyTorch, and Langchain. Proficient in Python and C++. Strong understanding of computer vision and deep learning. Passionate about building AI systems that solve real-world problems.

Skills

  • Programming Languages: Python, C++
  • AI/ML Frameworks: TensorFlow, PyTorch, Langchain
  • Deep Learning: CNNs, RNNs, Transformers
  • Computer Vision: Object detection, image classification, image segmentation
  • LLMs: GPT-4, GPT-4o
  • Model Development: Model optimization, fine-tuning and evaluation
  • Vulnerability Remediation: CVE analysis, Security fixes
  • Tools: Git, Jira, Weaviate, OpenCV, OpenCL, OpenGL

Experience

Model Development Engineer | Instabase | Bengaluru, Karnataka, India | 2020 - Present

  • LLM Development & Optimization:
    • Architected and productionized the first version of RAG based document understanding platform - AIhub Instabase
    • Developed and implemented a new chunk-based retrieval approach for answering questions directly, resulting in a 0.66875 score on local tests
    • Designed and implemented a system for handling out-of-context questions in a multistep system prompt
    • Optimized the performance of the MS lite model by reducing Weaviate calls for chunk to page mapping
  • Vulnerability Remediation & System Upgrades:
    • Remediated vulnerabilities in the Langchain library
    • Upgraded the OpenAI and Langchain libraries to the latest versions
  • General Model Improvements:
    • Fixed various bugs and issues in AI models and systems, enhancing their stability and reliability

Data Scientist | ShareChat | Bengaluru, Karnataka, India | 2019 - 2020

  • Visual Effects Development:
    • Worked on creating visual effects based on audio cues, using neural networks for audio source separation and OpenGL for effects
    • Implemented visual effects on Android platform
  • Content Moderation:
    • Contributed to NSFW/SFW media classification using multimodal neural networks
    • Involved in data cleaning, model training, and deploying on Kubernetes

Senior Engineer | Ittiam Systems | Bengaluru, Karnataka, India | 2019

  • Computer Vision Applications:
    • Contributed to person/face detection using single-shot neural networks
    • Contributed to improvements in person tracking using deep learning based feature matching
  • Edge Computing & Tooling:
    • Explored new edge hardware infrastructures for neural network inference
    • Helped to create mask annotation tools using state of the art methods like DEXTR and SiamMask

Engineer | Ittiam Systems | Bengaluru, Karnataka, India | 2017 - 2019

  • Algorithm Optimization:
    • Involved in improvements to optical flow/background modelling both in terms of accuracy and time performance
  • GPU Programming:
    • Implemented OpenCL GPU kernel for distortion correction
  • Software Architecture:
    • Coded a new algorithm pipeline architecture based on JobQueue design which resulted in a significant speedup of the algorithm

Education

B.E. (Honors) in Electronics and Communication Engineering (ECE)
R. V. College of Engineering, Bangalore | 2017

Awards and Recognition