AI & Software Engineer
Bridging the gap between Machine Learning research and production-grade software.
Specializing in high-throughput pipelines, Computer Vision, and scalable automation.

• 1919   LeetCode Contest Rating (Top 3.4 % , Peak Rank: 189 ).
• 400+   Algorithmic problems solved, focusing on complex Graph theory and Heap structures.
• 21 s   Processing time for 100 minutes of video using optimized cross modal retrieval pipelines. See the Project here
• 2 Retail outlets   actively powered by custom built commercial automation suites.

• The Challenge: Build a privacy first, cloud independent semantic video search tool that matches or beats commercial platforms.
  
• Engineering Milestone: Optimized inference to process 100 minutes of video in 21 seconds on a single T4 GPU.
  
• The Solution :
  • Engineered a cross modal retrieval pipeline utilizing SigLIP and FAISS to map natural language queries directly to high dimensional frame embeddings.
  • I Developed a low latency (< 300ms) multimodal RAG pipeline by indexing temporal video embeddings in a Qdrant vector database. Enabled natural language video search using text embeddings.
  • I eliminated GPU dependencies by optimizing SigLIP inference for consumer CPUs , leveraging ONNXINT8 quantization, FFmpeg pipeline , and scene aware frame extraction to ensure high-speed performance.
  • To execute automated deployment of container images on Docker Hub, I implemented Github Actions. Also used Jenkins to automate local Unit Testing before every commit to the main branch.
  • Ensured modular microservice architecture to handle high concurrency requests, ensuring scalability and performance. Implemented centralized logging and health check endpoints to ensure reliability.
  • This architectural knowledge of indexing, vector databases, and high concurrency retrieval is directly transferable to text-based databases.

• The Challenge : Eliminate expensive Business API costs for retail marketing teams by replacing manual WhatsApp outreach.
  
• Impact : Actively operational across two retail outlets (Cake of the Day), cutting operational ov erhead to zero for marketing workflows.
  
• The Solution : Built and deployed a self-managed automation and customer database application.
  
  • Engineered an application that reduced manual marketing overhead by 90% (from 10 to 12 hours to 30 to 40 minutes) for a retail chain using PyAutoGUI automation scripts for message delivery.
  • Delivered an application currently operational across two retail outlets of the 'Cake of the Day' commercial chain.
  • Encrypted MySQL backend for secure management of high volume commercial records.

• Education : B.E. Computer Engineering (AI/ML) + Honors in Data Science ( Currently in Final Year - sem 7 )
• Certifications : Google IT Automation Professional, Google Data Analytics Professional
• Core Focus: Scalable Automation, System Observability, and High Throughput Pipelines.

• Career Management Committee (CMC) : Elected as Executive Board Member, driving professional development initiatives.
• Technical Author : Authored and published a technical article on optimizing QA Chatbots using TensorFlow Feedforward Networks (FFN) on Ready Tensor.
• Engineering Project Lead : Spearheaded and managed the end-to-end delivery of 5 academic engineering projects, directing cross functional student developer teams. (4 to 5 members)
• Lead Guitarist : Directed musical arrangements and performed as the lead guitarist across 3 high scale music festival events, coordinating live performances under high pressure environments.
• Academic Excellence : Maintained a continuous streak of First Class Distinction from pre-secondary education through my Bachelor's Degree in Engineering.

LinkedIn   |   LeetCode   |   Email