Durgesh Dutt Sinha

AIML Engineer & Full-Stack Developer

MCA (AIML) scholar at Sri Balaji University Pune building production-grade AI systems, multi-modal computer vision pipelines, and resilient full-stack web platforms with sub-50ms inference.

Target: AIML Engineer • GenAI Developer • Full-Stack Architect Pune, Maharashtra, India (Open to Relocation & Remote) Available Immediately for 2026 Opportunities

Download PDF Resume (ATS Optimized)View HTML Resume ⚡ Performance Center

1-Click Candidate Summary for Hiring Teams

Durgesh Dutt Sinha — AIML Engineer & Full-Stack Architect • Master of Computer Applications (AIML) at Sri Balaji University Pune (2025–2027) | BCA Graduate. • Fellowships: UNLOX® AI Program Fellow, Be10x AI Cohort Member. • Core Competencies: Autonomous AI Agents, Real-Time MediaPipe Computer Vision (<50ms latency), Next.js 16, TypeScript, Drizzle ORM, Scikit-Learn (K-Means/GMM/DBSCAN), AWS Cloud Pipelines. • Featured Platforms: - RoleRadar: AI career platform with 8-point ATS scanner & Google XYZ bullet optimizer. - AI Fitness Ecosystem: Edge pose estimation & biomechanical form validation. - MarketMatch-AI: Retail RFM segmentation & lookalike recommendation engine. • Contact: durgeshdsinha@gmail.com | https://github.com/OxDurgeshxO | https://www.linkedin.com/in/durgesh-dutt-s-4ba74924b

Top 3 Flagship Engineering Showcases

Selected platforms with full architecture specs, tradeoffs, and verified metrics

RoleRadar

Full Stack AI & Career Tech • Flagship Production SaaS

Read Full Case Study

RoleRadar is an enterprise-grade career intelligence platform evaluating candidate resumes across 16 modern tech roles with an automated 8-point ATS compatibility scanner, dynamic Google XYZ bullet optimizer, and personalized multi-week learning roadmaps.

Role Benchmarks

16 Tech Roles

ATS Readiness

8-Point Audit

Database Resilience

100% Zero-Crash

Inference Latency

< 180ms

Next.js 16.2React 19TypeScript 5.9Drizzle ORMPostgreSQL (Neon)

AI Fitness Ecosystem & Pose Architecture

Computer Vision & Wearable IoT • Enterprise Multi-Modal Spec

Read Full Case Study

A production-grade 9-page multi-modal fitness architecture utilizing Google MediaPipe 33-point skeletal landmark detection, trigonometric joint-angle vector kinematics, and web worker isolation to provide real-time form correction under 50ms latency.

Pose Inference

< 45ms Latency

UI Responsiveness

60 FPS Solid

Platform Scope

9 Core Modules

Camera Privacy

100% Client-Side

TypeScriptGoogle MediaPipe PoseWeb Workers APIWebSocketsCanvas 2D / WebGL

MarketMatch-AI

Applied AI & Customer Intelligence • Tier-S Showcase

Read Full Case Study

An enterprise customer segmentation and lookalike audience platform transforming 4,000 retail records into actionable marketing cohorts using log-transformed RFM feature engineering, K-Means ($K=5$), Gaussian Mixture Models (GMM), and DBSCAN anomaly detection.

Customer Cohort

4,000 Records

Cluster Quality

0.507 Silhouette

Audience Export

1-Click CSV

Inference Speed

< 10ms

Python 3.11Scikit-LearnStreamlitPlotly 3DPandas & NumPy