Skip to content
AP
AI/ML Engineer

Arjun Pramod

AI/ML Engineer with hands-on experience building and deploying production systems in Computer Vision, Generative AI, and NLP. Strong in end-to-end model development, data pipelines, real-time inference, and cloud deployment using PyTorch, YOLO, LangChain, FastAPI, Docker, and AWS. Focused on shipping scalable AI solutions for practical business use.

System StatusLive
Status
AI Engineer @ Nesa Software
Base
Kerala, India
Focus
Computer Vision · Generative AI · NLP
Currently
AI Engineer @ Nesa Software
Education
B.Tech CSE (AI/ML), SRM University AP
Recognition
Best Paper Award — ICAIN 2025
Core stack
PyTorch · YOLO · LangChain · FastAPI · AWS
Selected Work

Featured Projects

Computer vision, applied ML, and generative AI systems, shipped end-to-end from data pipeline to deployed inference.

InteliCam

Suspicious Behavior Detection Platform

Mar 2026 – Present
  • Developed an AI-powered CCTV analytics platform for real-time and historical suspicious behavior detection, integrating YOLOv11 detection, pose estimation, ConvNeXt-Tiny classification, and XGBoost into a multi-stage inference workflow.
  • Built end-to-end data engineering pipelines with perceptual hashing (pHash) and clustering-based stratification; implemented Grad-CAM explainability and batch video inference workflows producing per-frame annotated output videos.
  • Developed RTSP recording, live inference, and video playback applications using FastAPI, OpenCV, FFmpeg, React, and Tailwind CSS for real-time CCTV monitoring, video playback, and interactive video analytics.
PyTorchConvNeXtOpenCVFastAPIReactTailwind CSS

DoorSense AI

Intelligent Door State Monitoring System

May 2026 – Present
  • Built an end-to-end CCTV analytics solution for automated door state classification (Open, Closed, Semi-Closed) from DVR/NVR streams, integrating YOLO-based detection with ConvNeXt-Tiny classification in a multi-stage real-time inference pipeline.
  • Collected and annotated surveillance data with Roboflow; engineered an automated ROI extraction pipeline using detector outputs with configurable padding to generate large-scale classification datasets.
  • Built a production-ready FastAPI web application supporting RTSP validation, DVR/NVR integration, inference configuration, monitoring dashboards, and full pipeline deployment.
PyTorchYOLOConvNeXt-TinyRoboflowFastAPIFFmpeg

Customer Retention Decision Platform

Applied Machine Learning System

Feb 2026 – Feb 2026
  • Built an end-to-end ML platform predicting telecom customer churn with XGBoost, capturing 72.5% of churners while targeting only 35.5% of the user base.
  • Developed an explainable AI system using SHAP and customer segmentation to generate persona-based retention recommendations served via a FastAPI endpoint.
  • Deployed on AWS EC2 with Docker containerization, a Streamlit monitoring dashboard, MLflow experiment tracking, and PostgreSQL prediction logging.
Scikit-learnXGBoostSHAPPostgreSQLMLflowFastAPIStreamlitDockerAWS

Retrieval-Augmented Document QA System

Applied NLP / RAG

Jan 2026 – Jan 2026
  • Built a Retrieval-Augmented Generation (RAG) API for document-grounded question answering using LangChain, FAISS vector search, and Sentence Transformer embeddings.
  • Implemented semantic search with FAISS and Maximal Marginal Relevance (MMR) retrieval to reduce hallucinations and return answers with traceable source citations.
  • Dockerized and deployed the FastAPI service on AWS EC2; diagnosed and resolved out-of-memory (OOM) issues during local LLM inference, improving production stability.
LangChainFAISSHugging FaceSentence TransformersFastAPIDockerAWS
Growth Timeline

Journey

From classroom fundamentals to production AI systems — the path so far.

Senior Secondary (Class X)

Jun 2018 – Mar 2019

Bhavans Adarsha Vidyalaya

Senior Secondary (Class XII)

Jun 2020 – Jul 2021

Saraswati Vidyaniketan Public School

B.Tech in Computer Science and Engineering (AI/ML specialization)

Jul 2021 – Nov 2025

SRM University AP

Artificial Intelligence Intern

Jul 2022 – Sep 2022

Zebo.ai

ML Research Intern

Jul 2023 – Sep 2023

SRM University AP

Best Paper Award, ICAIN 2025

2025

Lead author, ESADN research paper

AI Engineer

Mar 2026 – Present

Nesa Software Pvt. Ltd.

Current
Capabilities

Skills

Grouped by where they show up in the stack — from data and modeling through to deployment.

Programming & Databases

Languages and data stores behind every pipeline.

PythonSQLPostgreSQL

Machine Learning & Data

Classical ML and the data-science workflow that feeds it.

Scikit-learnXGBoostSHAPFeature EngineeringHyperparameter TuningModel EvaluationEDAData AnalyticsData ScienceNumPyPandasMatplotlibSeaborn

Deep Learning & Computer Vision

Training and deploying vision models, end to end.

PyTorchTensorFlowCNNsNeural NetworksTransfer LearningOpenCVYOLO (YOLOv11)ConvNeXtRoboflowObject DetectionImage ClassificationPose EstimationGrad-CAM

Generative AI & NLP

LLM-powered retrieval, agents, and language understanding.

LLMsLangChainTransformersHugging FaceSentence TransformersFAISSNLTKspaCyRAGAI AgentsNatural Language Processing (NLP)

Deployment & MLOps

Shipping models as reliable, observable services.

FastAPIREST APIsDockerAWS (EC2, S3)MLflowStreamlitGitFFmpegModel DeploymentReal-Time InferenceWebhooksn8n

Software Development

Application engineering around the AI core.

HTMLCSSJavaScriptReactTailwind CSSPyQtMultithreading
Track Record

Experience

Roles where models had to leave the notebook and run in production.

AI Engineer

Nesa Software Pvt. Ltd. · Kerala, India

Current role
Mar 2026 – Present
  • Architected and deployed production-grade computer vision systems for CCTV surveillance analytics, enabling automated real-time and historical suspicious activity detection across live DVR/NVR video streams.
  • Engineered an end-to-end door state monitoring pipeline integrating YOLO-based detection with ConvNeXt-Tiny multiclass classification (Open, Closed, Semi-Closed), covering Roboflow annotation, automated ROI extraction, and multi-stage real-time inference.
  • Built scalable data engineering pipelines with perceptual hashing (pHash) and clustering-based stratification for leakage-free splits; trained, evaluated, and deployed YOLOv11, ConvNeXt-Tiny classifiers, and XGBoost models for surveillance video analytics workflows.
  • Built and deployed a full-stack AI video analytics platform using Python, FastAPI, React, and Tailwind CSS, supporting RTSP stream ingestion, DVR/NVR integration, real-time AI inference, and video playback.
  • Designed and integrated LLM-powered Generative AI workflows — including retrieval-augmented generation (RAG) pipelines and AI-assisted automation features — for internal business applications.

ML Research Intern

SRM University AP · Andhra Pradesh, India

Jul 2023 – Sep 2023
  • Developed a preprocessing and augmentation pipeline (Python, OpenCV, Keras) generating 224×224 inputs; designed and trained a custom CNN over 15 epochs, achieving 95% validation accuracy and 92.63% test accuracy.
  • Evaluated the binary classifier with Scikit-learn metrics (macro F1: 0.93, confusion matrix) and packaged model artifacts for reproducible inference deployment.

Artificial Intelligence Intern

Zebo.ai · Bengaluru, India

Jul 2022 – Sep 2022
  • Built a DNN for heart disease prediction and a CNN-based dog breed classifier using TensorFlow and Keras transfer learning with data augmentation, covering end-to-end pipelines from preprocessing through model evaluation.
  • Improved model performance through systematic hyperparameter tuning and batch inference experiments, benchmarking configurations to identify optimal model architectures.
Recognition

Achievements

Moments that got flagged outside the day-to-day work.

Best Paper Award — ICAIN 2025

Lead author of “ESADN: Enhanced Spatial Attention Network for Road Accident Detection.”

Gold Medalist

1st place, ProductKraft Expo 1.0.

Certifications

Artificial Intelligence, Database Management Systems.

Foundations

Education

SRM University AP

B.Tech in Computer Science and Engineering (AI/ML specialization)

Andhra Pradesh, India · CGPA: 7.88

Jul 2021 – Nov 2025

Saraswati Vidyaniketan Public School

Senior Secondary (Class XII)

Kochi, Kerala · CBSE: 90.4%

Jun 2020 – Jul 2021

Bhavans Adarsha Vidyalaya

Senior Secondary (Class X)

Kochi, Kerala · CBSE: 92.2%

Jun 2018 – Mar 2019
Get in Touch

Reach out about Computer Vision, GenAI, or NLP work.

Based in Kerala, India.

arjunpramod509@gmail.com+91 9746782585linkedin.com/in/arjun-pramodgithub.com/ArjunPramod
Download full résumé