Skip to content
View GOKULRAM-K's full-sized avatar
4 followers · 6 following

Achievements

Achievement: Pull Sharkx3

Achievements

Achievement: Pull Sharkx3

Block or report GOKULRAM-K

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
GOKULRAM-K/README.md

Gokul Ram Kannan — Software and Machine Learning Engineer

Typing SVG



Portfolio   LinkedIn   Google Scholar   ORCID   Email

⚡ SPECIALIZATION  |  Production AI/ML • Distributed Backends • Real-Time Telemetry & Data Pipelines

📌 Executive Summary

Software & Machine Learning Engineer with proven experience designing, deploying, and optimizing end-to-end intelligent systems. Blends deep learning research with robust systems programming—scaling asynchronous backends, distributed streaming telemetry, and edge-to-cloud deployments.

  • Scale & Engineering Leadership: Global Rank #80 (Top 1% Globally of 43,587+ developers) in GSSoC 2026 with 179 merged pull requests across full-stack and ML production codebases.
  • Mission-Critical Systems: SIH 2025 Semi-Finalist for architecting an edge-intelligent dynamic power grid framework handling 604k+ streaming telemetry points with sub-50ms latency.
  • R&D and IP: 5 Issued Patents in sensor-grid telemetry and biofeedback loops, alongside 7 peer-reviewed publications (37+ citations) across Elsevier (Results in Engineering, IF 9.4 and Array, I.F. 5.3) and IEEE conferences.

🛠️ Technical Competencies

Domain Technologies & Tooling
Languages Python, C++, Java, TypeScript, JavaScript, SQL, C, Bash
AI / Machine Learning PyTorch, TensorFlow, Vision Transformers (ViT), GNNs, Hugging Face, OpenCV, Scikit-Learn, ONNX
GenAI & Vector Search Local LLMs (Ollama, vLLM), LangChain, LlamaIndex, FAISS, ChromaDB, RAG Pipelines
Backend & Distributed Systems FastAPI, Flask, Node.js, Next.js, REST APIs, WebSockets, Apache Kafka, Redis, Microservices
Data & Cloud Infrastructure PostgreSQL, MySQL, AWS (Solutions Architect & Cloud Practitioner Certified), Docker, Linux, CI/CD, Telemetry Pipelines

🚀 Featured Engineering Systems

🚗 DriveGraph

Autonomous Driving Simulator & XAI Telemetry

  • Coupled CARLA autonomous simulation with spatial-temporal Graph Neural Networks (GNNs) for predictive collision risk assessment.
  • Processed real-time vehicle kinematics telemetry, multi-agent trajectories, and attention-based explainability heatmaps.

Python PyTorch CARLA Simulator GNN OpenCV FastAPI

⚡ EdgeGrid-AI

Smart Grid Dynamic Phase Balancing (SIH Finalist)

  • Architected edge-intelligent framework processing 604k+ telemetry points with sub-50ms relay switching latency.
  • Deployed Kafka message streaming across Raspberry Pi edge nodes, cutting phase unbalance by 37%.

Apache Kafka Docker Raspberry Pi Edge AI Python React

⚖️ LexBNS

Offline High-Throughput RAG Architecture

  • Engineered a fully offline, zero-data-leakage legal retrieval engine powered by quantized Llama 3.1 via Ollama.
  • Implemented FAISS vector indexing with hierarchical document chunking, achieving sub-800ms query retrieval latency.

Python Ollama FAISS LangChain FastAPI Docker

🛡️ DeepShield

Vision Transformer Video Forensics Pipeline

  • Implemented frame-level video forensics engine using Vision Transformers (ViT), achieving 87.8% validation accuracy.
  • Built asynchronous inference service with automated face extraction, landmark alignment, and spatial heatmaps.

PyTorch Vision Transformers OpenCV Flask Inference Optimization

🌐 Open-Source Contributions & Production Code

GLOBAL RANK
#80
Top 1% of 43,587+ Developers 		PRODUCTION MERGED PRs
179
Full-Stack & ML Contributions 		LEADERBOARD SCORE
28,813
GirlScript Summer of Code 2026
  • pathfinder-ai (91 Merged PRs): Built high-throughput resume parsing pipelines, asynchronous multi-format OCR extraction (PDF/DOCX), and FAISS semantic matching.
  • bizinsight-ai: Optimized PostgreSQL schemas, implemented Redis caching layers, and decoupled backend analytics endpoints.
  • logara-ai: Refactored training telemetry pipelines, epoch-level metric normalization, and automated CI tests.

💡 Patents & Research Publications

  • 🌾 AI-Driven Smart Irrigation Controller & Telemetry System (App No: 202541038729 • Patent Issued, 2025)
    Automated sensor-grid telemetry, soil moisture predictive models, and cloud-actuated micro-drip optimization.
  • 💖 Affective Telemetry & Real-Time Biometric Feedback Loop (App No: 202441052499 • Patent Issued, 2024)
    Real-time facial micro-expression telemetry and physiological biofeedback architecture.
  • 📄 Peer-Reviewed Literature: 7 publications with 37+ citations across Elsevier (Results in Engineering, IF 9.4) and IEEE on explainable AI, audio DSP, and distributed swarm systems.

📊 Contribution Activity

Contribution Snake

Portfolio  •  LinkedIn  •  Google Scholar  •  ORCID  •  GSSoC Profile  •  Email

Engineered with precision, scalable architecture, and production reliability.

Pinned Loading

  1. DeepShield DeepShield Public

    This project detects deepfake videos using a Vision Transformer (ViT) model, classifying frames as real or manipulated with high accuracy.

    JavaScript

  2. EdgeGrid-AI EdgeGrid-AI Public

    This project builds a self-learning, hybrid IoT architecture that connects every household node and transformer to a central intelligence layer — providing real-time visibility, load balancing, and…

    Python

  3. English-Speech-Emotion-Recognition-Framework English-Speech-Emotion-Recognition-Framework Public

    Cross-corpus and noise-robust speech emotion recognition using a lightweight 1D CNN across RAVDESS, CREMA-D, TESS, and SAVEE.

    Jupyter Notebook

  4. Legal-RAG-Agent Legal-RAG-Agent Public

    Reasoning-enhanced legal AI assistant combining ReAct, RAG, FAISS, and LLMs to query Indian legal documents.

    Python

  5. ProteinFold-XAI ProteinFold-XAI Public

    Engineered an ensemble deep learning pipeline for protein secondary structure prediction, integrating BiLSTM architectures, NLP-based evaluation, and XAI (LIME, Integrated Gradients). Achieved 94.4…

    Jupyter Notebook

  6. SamvidhaanAI SamvidhaanAI Public

    A legal chatbot designed to improve accessibility to the Indian Constitution using a Retrieval-Augmented Generation (RAG) pipeline powered by Google's Gemini AI.

    Jupyter Notebook