ga★
Garvit AroraPORTFOLIO 2026 — AI & SOFTWARE
Born to build. Raised by deadlinesga★An AI Engineer & Builder with a heart full of caffeineGARVIT.AI ★ 2026
Born to build. Raised by deadlinesga★An AI Engineer & Builder with a heart full of caffeineGARVIT.AI ★ 2026
Born to build. Raised by deadlinesga★An AI Engineer & Builder with a heart full of caffeineGARVIT.AI ★ 2026
UNBOXINGMY........
Born to buildga★GARVIT ARORA // 2026—27ga★
HIGH VALUE !!!AI × SOFTWARE !!!9.11 CGPA !!!
FRAGILE // HANDLE WITH LOVE
PRIORITY PARCEL★ 2026/27

SHIPPING PORTFOLIO

FROM: GARVIT ARORA (SRMIST KTR)

GARVIT.AI ★
ORGANIZED CODE
BORN TO BUILD. RAISED BY DEADLINES.
Born to build. Raised by deadlinesga★An AI Engineer & Builder with a heart full of caffeineGARVIT.AI ★ 2026
Born to build. Raised by deadlinesga★An AI Engineer & Builder with a heart full of caffeineGARVIT.AI ★ 2026
Born to build. Raised by deadlinesga★An AI Engineer & Builder with a heart full of caffeineGARVIT.AI ★ 2026
Born to create,
Raised by deadlines
welcome to my portfolio
With a heart
full of AI and caffeine
AI Engineerand Software Developer

PORTFOLIO

ga★GARVIT.
Garvit Arora peeking into the portfolio box
20
26—27
I BUILD THINGS THAT THINK. — SRMIST KTR (CGPA 9.11/10)
Where technology
meets business
★✿✦
Garvit Arora — AI Engineer and Software Developer
✆
✕
LANGUAGE

ENGLISH & HINDI:Professional working & native proficiency

EDUCATION

B.Tech in Computer Science and Business Systems

SRM Institute of Science and Technology, Kattankulathur

CGPA: 9.11/10 (Class of 2027)

HARD SKILLS
PyPython
PtPyTorc
CvOpenCV
TsTypeSc
NxNext.j
SqSQL /
My GitHub:GarvitPro
EXPERIENCE
Oct 2026 – Present
AI Engineer Internat INVENTIC.AI (Remote)
June 2026 – July 2026
AI & Computer Vision Internat DRDO — IRDE (Dehradun, Uttarakhand)
Dec 2025 – Jan 2026
Product & Systems Internat GREENHECK (Gurugram, Haryana)
June 2025
Operations & Data Internat SAMBANDH (Gurugram, Haryana)
CERTIFICATE
2026
Google Data AnalyticsIssuer: Coursera (June 2026)
2026
Salesforce Tableau Desktop FoundationIssuer: Salesforce (March 2026)
2025
Advanced SQL & DBMSIssuer: Scaler (May 2025)
2026
SAP Generative AI DeveloperIssuer: SAP (March 2026 – March 2027)
SOFT SKILLS
  • Analytical Problem-Solving
  • Systems & Business Thinking
  • Technical Research & Experimentation
  • Root-Cause Gap Analysis
  • Cross-Functional Collaboration
  • Rapid Full-Stack Prototyping
HIGHLIGHTS
2026
DRDO — IRDE Defence Research

Underwater periscope image restoration & super-resolution pipeline.

2026
Self-Pruning Neural Network

33.42% weight sparsity & 54.19% accuracy on CIFAR-10 at λ = 1e-4.

OBJECTIVE

I'm Garvit Arora, a Computer Science & Business Systems undergraduate at SRMIST. My goal is to engineer intelligent AI systems and full-stack software that bridge deep technical research with real-world business impact.

INTERESTS
  • Deep Learning Architectures
  • Computer Vision & Restoration
  • Enterprise ERP & Decision Logic
  • Hackathons & Product Building
  • Financial ML & Data Journalism
  • Late-Night Coffee & Shipping Code
let's unbox this first.
01
4× INTERNSHIPS // INVENTIC.AI · DRDO · GREENHECK · SAMBANDH

EXPERIENCE

Inventic.ai

AI Engineer Intern — October 2026 – Present (Remote)

BRIEF

Contributing to the development and implementation of AI-based solutions for practical applications and real-world challenges. Exploring and applying emerging artificial intelligence technologies.

ENGINEERING IDEA★

Working on AI projects requiring technical problem-solving, experimentation, and iterative improvement. Gaining practical experience in a remote AI engineering environment.

TOOLS USED

Artificial Intelligence · Applied AI Engineering · Technical Experimentation · Iterative Prototyping · Problem-Solving

PROJECT TYPE

*AI Engineer Intern
*Remote (October 2026 – Present)
INTERNSHIP WORKFLOW & DELIVERABLES // INVENTIC.AI
REAL-WORLD BRIEFAI PROTOTYPINGEXPERIMENTATIONMODEL EVALUATIONITERATIVE TUNINGAPPLIED SOLUTION
01Contributing to the development and implementation of AI-based solutions for practical applications and real-world challenges.
02Exploring and applying emerging artificial intelligence technologies.
03Working on AI projects requiring technical problem-solving, experimentation, and iterative improvement.
04Gaining practical experience in a remote AI engineering environment.
October 2026 – PresentRemote

Defence Research and Development Organisation — IRDE

AI & Computer Vision Intern — June 2026 – July 2026 (Dehradun, Uttarakhand)

BRIEF

Developed a Python-based image-processing pipeline to preprocess and enhance underwater periscope imagery. Evaluated image restoration and super-resolution techniques under varying image conditions.

ENGINEERING IDEA★

Compared experimental outputs to assess processing quality and validate workflows. Identified suitable image-enhancement approaches through structured experimentation. Collaborated with research team members to translate technical requirements into image-processing workflows.

TOOLS USED

Python · Image Processing · Computer Vision · Image Restoration · Experimental Evaluation · Technical Research

PROJECT TYPE

*AI & Computer Vision Intern
*Dehradun, Uttarakhand (June 2026 – July 2026)
INTERNSHIP WORKFLOW & DELIVERABLES // DRDO — IRDE
STAGE 01 // PREPROCESS

Underwater Periscope Input

Python pipeline handling noise, turbidity & contrast degradation.

STAGE 02 // RESTORATION

Enhancement & Super-Resolution

Structured experimentation across varying underwater conditions.

STAGE 03 // VALIDATION

Comparative Quality Audit

Benchmarking outputs with DRDO — IRDE research team members.

01Developed a Python-based image-processing pipeline to preprocess and enhance underwater periscope imagery.
02Evaluated image restoration and super-resolution techniques under varying image conditions.
03Compared experimental outputs to assess processing quality and validate workflows.
04Identified suitable image-enhancement approaches through structured experimentation.
05Collaborated with research team members to translate technical requirements into image-processing workflows.
June 2026 – July 2026Dehradun, Uttarakhand

Greenheck Pvt. Ltd.

Product & Systems Intern — December 2025 – January 2026 (Gurugram, Haryana)

BRIEF

Analyzed SAP ERP workflows and applied rule-based validation logic to support data consistency. Performed root-cause analysis of Bill of Materials configuration issues.

ENGINEERING IDEA★

Identified recurring data-validation gaps and workflow issues. Designed and tested decision logic for manufacturing configuration scenarios. Collaborated with engineering stakeholders to translate business requirements into structured validation workflows.

TOOLS USED

SAP ERP · Enterprise Systems · Business Process Analysis · BOM · Data Validation · Root Cause Analysis · Rule-Based Systems

PROJECT TYPE

*Product & Systems Intern
*Gurugram, Haryana (December 2025 – January 2026)
INTERNSHIP WORKFLOW & DELIVERABLES // GREENHECK
01 // SAP ERP WORKFLOW

Bill of Materials (BOM)

Analyzed manufacturing configuration scenarios & validation gaps.

02 // ROOT-CAUSE RCA

Decision Logic Design

Engineered rule-based checks to eliminate recurring config errors.

03 // DATA CONSISTENCY

Stakeholder Alignment

Translated engineering requirements into structured ERP rules.

01Analyzed SAP ERP workflows and applied rule-based validation logic to support data consistency.
02Performed root-cause analysis of Bill of Materials configuration issues.
03Identified recurring data-validation gaps and workflow issues.
04Designed and tested decision logic for manufacturing configuration scenarios.
05Collaborated with engineering stakeholders to translate business requirements into structured validation workflows.
December 2025 – January 2026Gurugram, Haryana

Sambandh Health Foundation

Operations & Data Intern — June 2025 (Gurugram, Haryana)

BRIEF

Cleaned, transformed, and organized data collected from multiple sources. Improved data quality for reporting and operational analysis.

ENGINEERING IDEA★

Conducted exploratory data analysis. Investigated data inconsistencies and reporting bottlenecks using root-cause analysis. Supported structured reporting and data-informed operational decision-making.

TOOLS USED

Data Cleaning · EDA · Data Quality · Operational Analytics · Reporting · Root Cause Analysis

PROJECT TYPE

*Operations & Data Intern
*Gurugram, Haryana (June 2025)
INTERNSHIP WORKFLOW & DELIVERABLES // SAMBANDH
01 // MULTI-SOURCE DATA

Cleaning & Transformation

Standardized raw operational feeds for accurate reporting.

02 // EXPLORATORY EDA

Bottleneck Investigation

Root-cause analysis of data inconsistencies & delays.

03 // DECISION SUPPORT

Operational Analytics

Structured reporting enabling data-informed leadership decisions.

01Cleaned, transformed, and organized data collected from multiple sources.
02Improved data quality for reporting and operational analysis.
03Conducted exploratory data analysis.
04Investigated data inconsistencies and reporting bottlenecks using root-cause analysis.
05Supported structured reporting and data-informed operational decision-making.
June 2025Gurugram, Haryana
let's open the next one.
02
PYTORCH · COMPUTER VISION · IMBALANCE-AWARE ML

AI & DEEP LEARNING

SELF-PRUNING NEURAL NETWORK

Deep Learning / AI Research — Learnable Gating & Sparsity Regularization on CIFAR-10

BRIEF

Neural networks can contain many parameters. This experiment investigates whether learned gating can increase sparsity while retaining useful predictive performance.

ENGINEERING IDEA★

Build a custom gate-controlled neural network in PyTorch using custom PrunableLinear layers and evaluate different sparsity penalties across controlled training runs.

TOOLS USED

Python · PyTorch · Neural Networks · CIFAR-10 · Sparsity Regularization

PROJECT TYPE

*Deep Learning / AI Research
*Documented Research Experiment
ARCHITECTURE & EXPERIMENT BOARD // SELF-PRUNING NEURAL NETWORK
ARCH // PRUNABLE-LINEAR FEED-FORWARD (CIFAR-10)W_eff = W ⊙ σ(g) | L1 PENALTY λ = 1e-4
INPUT xGATE σ(g₁)GATE σ(g₂)LOGITS ŷDOCUMENTED REPORTCONFIG: λ = 1e-4WEIGHT SPARSITY33.42%TEST ACCURACY (CIFAR-10)54.19%--- PRUNED GATE (σ(g) → 0)─── ACTIVE WEIGHT CONNECTION
Regularization Strengthλ = 1e-4Documented report benchmark configuration
Test Accuracy54.19%Evaluated on CIFAR-10 test split at λ = 1e-4
Achieved Sparsity33.42%Pruned weight connections documented in report
PROCESS & IMPLEMENTATION STAGES
STAGE 01

CIFAR-10 Input Vectorization

32×32×3 RGB image tensors flattened and normalized for feed-forward classification.

STAGE 02

PrunableLinear Gated Weights

Each weight matrix W is paired with learnable gate parameters g, producing effective weights via sigmoid gating σ(g).

STAGE 03

Joint Cross-Entropy + L1 Gate Penalty

Loss function combines classification loss with λ · ||σ(g)||₁ to drive non-essential connection gates toward zero.

STAGE 04

Threshold Pruning & Evaluation

At λ = 1e-4, documented test accuracy reaches 54.19% while achieving 33.42% weight sparsity.

FIG 04.01 — ILLUSTRATIVE DIAGRAM OF LEARNABLE GATE PRUNING METHOD (W_eff = W ⊙ σ(g)); METRICS REFLECT DOCUMENTED λ = 1e-4 EXPERIMENT.

CREDIT CARD FRAUD DETECTION

Machine Learning / Financial Technology — Imbalance-Aware Classification on Transactional Data

BRIEF

Fraudulent transactions are uncommon compared with legitimate transactions, making classification challenging and rendering naive accuracy metrics misleading.

ENGINEERING IDEA★

Prepare transaction data, address class imbalance with SMOTE, train ensemble classification models, and evaluate performance with precision, recall, F1-score, ROC-AUC, and confusion matrices.

TOOLS USED

Python · Scikit-learn · Random Forest · SMOTE · Pandas · NumPy

PROJECT TYPE

*Machine Learning / Financial Technology
*Machine Learning Experiment
ARCHITECTURE & EXPERIMENT BOARD // CREDIT CARD FRAUD DETECTION
PIPELINE // SMOTE + RANDOM FOREST CLASSIFIERMETRICS: PRECISION · RECALL · F1 · ROC-AUC
FEATURE SPACE PARTITION // SMOTE MINORITY RESAMPLING● MAJORITY (LEGITIMATE) ● FLAGGED ANOMALY ○ SMOTE SYNTHETIC SAMPLEEVALUATION FRAMEWORKTRUE POS (TP)Recall FocusFALSE POS (FP)Precision CostFALSE NEG (FN)Critical RiskTRUE NEG (TN)SpecificityPRIMARY METRICS:ROC-AUC · F1-SCORE · PR-CURVE
PROCESS & IMPLEMENTATION STAGES
STAGE 01

Exploratory Data Analysis & Scaling

Inspect feature distributions, normalize transaction amount and time variables, and isolate class skew.

STAGE 02

SMOTE Minority Resampling

Synthesize balanced training distributions in feature space to prevent majority-class dominance.

STAGE 03

Random Forest & Model Comparison

Fit ensemble tree classifiers and benchmark against baseline classification approaches.

STAGE 04

Imbalance-Aware Evaluation

Analyze Confusion Matrices, Precision-Recall balance, F1-score, and ROC-AUC curves.

FIG 04.03 — ILLUSTRATIVE ANIMATION OF CLASS-IMBALANCE CLUSTERING, SMOTE RESAMPLING & DECISION BOUNDARY PARTITIONING.

DEEPSAFE

Computer Vision / Deep Learning — Deepfake Detection Research Prototype

BRIEF

Synthetic facial media requires structured preprocessing pipelines that can isolate facial regions consistently before feeding frames into convolutional classification architectures.

ENGINEERING IDEA★

Build a modular Python pipeline combining OpenCV face localization and cropping with a TensorFlow CNN training architecture, paired with an interactive Streamlit prototype interface.

TOOLS USED

Python · OpenCV · TensorFlow · Streamlit · NumPy

PROJECT TYPE

*Computer Vision / Deep Learning
*Research Prototype (Training Pipeline + Demo UI)
ARCHITECTURE & EXPERIMENT BOARD // DEEPSAFE
PROTOTYPE // OPENCV FACE EXTRACTION + TENSORFLOW CNNSTATUS: RESEARCH PROTOTYPE
OPENCV ROI EXTRACTION // ABSTRACT MESH01 // TRAINING & PREPROCESSING PIPELINE• OpenCV Face Bounding Box & Crop• NumPy Normalization & Tensor Resizing• TensorFlow CNN Feature Extraction Code02 // STREAMLIT DEMONSTRATION INTERFACE• Interactive Media Upload & Inspection UI• NOTE: Demo UI generates simulated predictions• Documented honestly as a research prototype
PROCESS & IMPLEMENTATION STAGES
STAGE 01

Media Ingestion & Frame Decoding

Load input imagery and convert color spaces into normalized NumPy arrays.

STAGE 02

OpenCV Face Localization & Cropping

Detect facial bounding boxes, extract aligned face regions, and resize tensors for CNN input.

STAGE 03

TensorFlow CNN Training Pipeline

Convolutional feature extraction layers and binary classification head coded in TensorFlow.

STAGE 04

Streamlit Demonstration Interface

Interactive prototype UI illustrating the user flow (currently operating with simulated demonstration predictions).

FIG 04.04 — ORIGINAL ABSTRACT WIREFRAME LANDMARK & PREPROCESSING SCHEMATIC. PROTOTYPE UI CURRENTLY USES SIMULATED DEMO OUTPUTS.
Let's turn it on !
03
NEXT.JS · TYPESCRIPT · POSTGRESQL · TABLEAU

FULL-STACK SYSTEMS

ATOMQUEST

Full-Stack / Enterprise Software — Goal Setting & Tracking Portal

BRIEF

Organizational goal tracking often fragments across spreadsheets and disconnected approval chains, making quarterly progress visibility and auditability difficult for employees, managers, and HR teams.

ENGINEERING IDEA★

Design a unified Next.js + TypeScript portal with strict Zod schema validation, role-based demonstration workflows, interactive Recharts analytics, and tabular export utilities.

TOOLS USED

Next.js · TypeScript · React · Tailwind CSS · NextAuth · Recharts · React Hook Form · Zod

PROJECT TYPE

*Full-Stack / Enterprise Software
*Hackathon Application & Interactive Demo
INTERFACE & SYSTEM ARCHITECTURE // ATOMQUEST
ATOMQUEST // ROLE-SCOPED GOAL PORTAL
NEXT.JS + ZOD + RECHARTS
01 / EMPLOYEE WORKFLOWGoal Creation & Zod FormQuarterly KPIs + Weights02 / MANAGER APPROVALReview & Sign-Off QueueAudit Log + Notifications03 / HR ADMIN & EXPORTAnalytics & CSV / XLSXOrg-Wide Quarterly TrackingQUARTERLY ACHIEVEMENT & APPROVAL ANALYTICS (RECHARTS LAYER)Q1Q2Q3Q4■ DEFINED OBJECTIVES ■ APPROVED & VERIFIED MILESTONES
STAGE 01

Role-Scoped Entry & Session Context

NextAuth-configured architecture supporting distinct views for Employees, Managers, and HR Administrators.

STAGE 02

Validated Goal Authoring

React Hook Form + Zod validation enforces structured objectives, quarterly milestones, and measurable weightings.

STAGE 03

Approval & Audit Pipeline

Manager review queues with status transitions, timestamped audit logs, and automated workflow notifications.

STAGE 04

Analytics & Tabular Exports

Recharts visual reporting paired with CSV/XLSX data export for administrative review.

FIG 04.02 — EDITORIAL INTERFACE SCHEMATIC OF THE HACKATHON DEMONSTRATION PORTAL (EMPLOYEE → MANAGER → HR WORKFLOWS).

FITNESS LIFESTYLE IMPACT ANALYZER

Full Stack / Data Visualization — Full-Stack Lifestyle Analytics & Visual Reporting

BRIEF

Raw lifestyle and activity logs are difficult to interpret in isolation without structured relational storage and clear visual reporting interfaces.

ENGINEERING IDEA★

Combine a type-safe React + TypeScript web interface and Node.js/Express + PostgreSQL backend with Tableau visual reporting dashboards.

TOOLS USED

React · TypeScript · Node.js · Express · PostgreSQL · Tableau

PROJECT TYPE

*Full Stack / Data Visualization
*Full-Stack Analytics Application
INTERFACE & SYSTEM ARCHITECTURE // FITNESS LIFESTYLE IMPACT ANALYZER
STACK // REACT · TYPESCRIPT · NODE/EXPRESS · POSTGRESQL · TABLEAUVISUAL REPORTING
RELATIONAL & API ARCHITECTUREReact + TypeScript FrontendTyped Lifestyle Input & Reporting ViewsNode.js + Express Backend APIREST Routing & Aggregation ControllersPostgreSQL Relational DatabaseStructured Lifestyle & Activity RecordsTABLEAU VISUAL REPORTING INTERFACEILLUSTRATIVE DASHBOARD SCHEMATIC // MULTIVARIATE TRENDS
STAGE 01

React & TypeScript Frontend

Structured input views and modular dashboard containers built with strongly typed components.

STAGE 02

Node.js & Express API Layer

Backend routing and controller logic handling data ingestion and query aggregation.

STAGE 03

PostgreSQL Relational Storage

Normalized relational schema managing structured lifestyle and activity attributes.

STAGE 04

Tableau Visual Reporting

Interactive analytical views designed to surface patterns and support data interpretation.

FIG 04.05 — ILLUSTRATIVE ANALYTICS DASHBOARD SCHEMATIC (REACT + NODE/EXPRESS + POSTGRESQL + TABLEAU REPORTING).
Let's listen to what it says...
04
6 SECTORS // RESUME & GITHUB PROJECT-VERIFIED SKILLS

TECH UNIVERSE

THE TOOLS BEHIND THE WORK

CLICK OR HOVER ANY SKILL STICKER TO SEE WHICH PROJECT OR INTERNSHIP PROVES IT ★
INSPECTING SKILL STICKER // PyTorchGITHUB PROJECT DEMONSTRATED
FEATURED PROJECT EVIDENCE:

★ Self-Pruning Neural Network

Open Project Case Study
INTERNSHIP / FIELD EVIDENCE:

SRMIST B.Tech CSBS Coursework & Labs

01 // Programming Languages
02 // Machine Learning & Artificial Intelligence
03 // Data Analytics
04 // Data Visualization
05 // Databases & Enterprise Systems
06 // Developer Tools & Project-Demonstrated Stack
Let's enjoy it together
05
SRMIST KATTANKULATHUR // B.TECH CSBS (CGPA 9.11/10)

EDUCATION

ACADEMIC FOUNDATION // CLASS OF 2027
BACHELOR OF TECHNOLOGY (2023 — 2027)

SRM Institute of Science and Technology

Computer Science and Business Systems (CSBS)

“My academic foundation combines computer science, analytical problem-solving, and business systems — an intersection that continues to shape how I approach technology.”

Kattankulathur, Tamil Nadu, India
CUMULATIVE GPA
9.11 / 10
EXPECTED GRADUATIONCLASS OF 2027
Let's feel the last one <3
06
COURSERA (GOOGLE) · SALESFORCE (TABLEAU) · SCALER · SAP

CERTIFICATIONS

CONTINUOUSLY LEARNING

VERIFIED TECHNICAL CREDENTIALS RECORDED ON RESUME
CREDENTIAL 01 // COURSERA
DATA ANALYTICS & PIPELINES

Google Data Analytics

Issued by: Coursera

June 2026
CREDENTIAL 02 // SALESFORCE
VISUAL INTELLIGENCE & BI

Salesforce Tableau Desktop Foundation

Issued by: Salesforce

March 2026
CREDENTIAL 03 // SCALER
RELATIONAL ARCHITECTURE

Advanced SQL & DBMS

Issued by: Scaler

May 2025
CREDENTIAL 04 // SAP
ENTERPRISE GENERATIVE AI

SAP Generative AI Developer

Issued by: SAP

March 2026 – March 2027Mar 2026 – Mar 2027
THE END
ga★

FROM A HEART FULL OF
DEADLINES, AI AND CAFFEINE

THANK YOU !!!

OPEN FOR AI & SOFTWARE ENGINEERING ROLES

LET'S BUILD SOMETHING MEANINGFUL.

I'm interested in opportunities involving artificial intelligence, software engineering, data-driven problem-solving, and technology-led business innovation. Whether you're a recruiter, an engineer, or someone working on an ambitious idea, I'd be glad to connect.