AI Engineer & Researcher
000%
Astrophage
Tau Ceti · 12 light-years · Entity unknown
click anywhere to return
AI Engineer & Researcher

Builder of
things that
actually ship.

I build production ML systems, publish research in Springer journals, and occasionally win hackathons. Mumbai → Nijmegen.

Currently → AI Developer, Vector Consulting, Mumbai
Next     → MSc Computer Science, Radboud University

RA 19h 47m 12s · DEC +26° 07′ · z = 0.00000 · J2026.3
career stats · 2021–2026
02
Springer paperspeer-reviewed research
01
Under reviewMultimedia Tools & Applications
12
LanguagesHisaab-Book · Bhashini Sprint
1K+
Students placedHead of Tech Ops · TPC, RAIT
3+
Years prod. MLlogistics · NLP · vision
ablaze@hail-mary:~/ai-engine — python3
What I do

Research. Build. Deploy.

Not in that order. Sometimes all three at once.

01
Build

Production ML pipelines for logistics, speech-to-task systems, LLM deployment, semantic tagging, CI/CD automation. If it runs in production and breaks at 2AM, I've probably shipped something exactly like it.

02
Research

Deepfake detection, computer vision, multimodal forensics. Two publications in Springer. VeriLens is my current work — a full detection framework evaluated on Celeb-DF v2 and DFDC, the hardest benchmarks available.

03
Solve

Hackathons, consulting, logistics AI. Bhashini Sprint finalist. HackThisFall winner. I work best when the problem is real and the deadline is violent. ClearPath Vision is what I'm building next.

Research

Published Work

I publish what I can't solve with just an experiment.

Springer · Under Review
VeriLens: A Deepfake Detection Framework Using VGG16, LRCN, and SVM

A multi-stage pipeline for automated deepfake detection: Haar Cascade face localization, VGG16 convolutional feature extraction, and parallel classification via LRCN for temporal sequence modeling and SVM for margin-based discrimination. Evaluated on Celeb-DF v2 and DFDC.

AbduRahim Shaikh, Yashas Kulkarni, Rahul Barna, Jamal Mydeen
Guide: Prof. Namita D. Pulgam · RAIT, Mumbai University
Target Journal
Multimedia Tools & Applications · Springer Nature
Architecture
Haar Cascade → VGG16 → LRCN / SVM ensemble
Datasets
Celeb-DF v2 · DFDC
Status
Two-round revision · active
Request Preprint →
Springer CCIS · Published 2021
VeriFace: Early Deepfake Detection Using Facial Feature Analysis

First publication. A deepfake detection system using facial feature analysis and classical ML classification. Published in Springer CCIS. The groundwork that became VeriLens.

AbduRahim Shaikh et al. · RAIT, Mumbai University
Series
Springer CCIS
Year
2021
Lineage
VeriFace → VeriLens
View Publication →
Selected Work

Projects

Three things I'm most proud of. Full work on GitHub.

Startup · In Development
ClearPath Vision

My own venture. AI systems for logistics and supply chain intelligence — building on two years of production experience with warehouse tracking, ETA prediction, and shipment anomaly detection.

MLOpsLogistics AIETA PredictionAnomaly Detection
RL · Systems
Autonomous Driving Agent

Deep Q-Learning agent trained from scratch in a simulated driving environment. Full RL pipeline — reward engineering, environment design, policy optimization. Zero to stable policy.

Deep Q-LearningTensorFlowReward EngineeringPython
Stack

What I work with

The tools I reach for when the problem is real.

Frameworks
TensorFlowPyTorchscikit-learnOpenCVKeras
Languages
PythonSQLJavaJavaScript
Deployment
FlaskDockerCI/CDREST APIsGit
NLP & Speech
WhisperBhashini APILLM DeploymentSemantic TaggingRVC Pipeline
Databases
PostgreSQLMySQLRedis
Research
LaTeXJupyterCeleb-DF v2DFDCAcademic Writing
Timeline

The short version.

Every entry that earned its place.

2026
VeriLens — Springer MTTA Revision Active
First-author paper. Full reviewer response, two-round revisions. Submitted to Multimedia Tools and Applications, Springer Nature.
2025
Admitted — MSc Computer Science, Radboud University Upcoming
Nijmegen, Netherlands. One of the strongest CS programmes in the Netherlands. Moving from Mumbai.
2024
AI Developer — Vector Consulting Group, Mumbai
Building production ML systems: LLM deployment, speech-to-task pipelines, logistics AI, warehouse automation. Real systems, real stakes.
2023
Bhashini Sprint Finalist Top 10
Hisaab-Book — 12-language AI inventory management. National-level government AI hackathon.
2023
HackThisFall — Best Project Idea
Recognised nationally for innovation and product thinking. Competed against 400+ teams.
2023
B.Tech Computer Engineering (AI) — RAIT, Mumbai University
Ramrao Adik Institute of Technology. Specialisation in Artificial Intelligence.
2022
AI Research Intern — AjnaLens
Applied computer vision research at an AR/AI hardware company. First exposure to production vision pipelines in hardware.
2021
VeriFace — Springer CCIS Series Published
First publication. Deepfake detection using facial feature analysis. The paper that started everything.
Mumbai → Nijmegen

AbduRahim
Shaikh

I'm an AI Engineer based in Mumbai. I build ML systems that go into production — pipelines for logistics, deepfake detectors, speech interfaces — and I publish the harder stuff as research.

I like problems that are real. Not Kaggle real. Not demo real. Production real, where a bug at 2AM matters and someone's business depends on the output.

I went to school in Yanbu, Saudi Arabia, did my engineering at RAIT in Mumbai, and I'm moving to the Netherlands for a master's at Radboud University.

Outside of code, I write poetry — the emotionally messy kind. I listen to Joji, Glass Animals, Billie Eilish. I play chess badly and lose gracefully.

PoetryChess Joji · Glass AnimalsGaming Atmospheric MusicIndie Sci-fi
Let's build
something.

Open to research collaborations, ML consulting, and conversations about production AI systems. I respond to real asks.