U K Rithika
Portfolio · 2024 / 25 · NIFT Chennai

Hi, I'm Rithika.

Engineeringthe Futureof Apparel.

Fashion Technology Graduate · Apparel Production
AI · IoT · Smart Manufacturing

Explore Projects→Let's Connect
[01]About
U K Rithika

Where the cutting floor meets the algorithm.

I'm a Fashion Technology graduate from NIFT Chennai specialized in Apparel Production — with a sharpened eye for smart manufacturing systems, AI-driven process optimization, apparel analytics, lean implementation, IoT applications and product development. Her practice fuses craft with data, engineering quality and efficiency into every stitch of the modern fashion supply chain.

07+
Live projects
85%
Model accuracy
3s
Defect alert
Apparel ProductionAI in FashionSmart ManufacturingProduct DevelopmentIoT ApplicationsLean SystemsFashion Technology
Apparel Production·AI in Fashion·Smart Manufacturing·IoT·Lean Systems·Data Analytics·Apparel Production·AI in Fashion·Smart Manufacturing·IoT·Lean Systems·Data Analytics·
[02]Capabilities

A toolkit built for the next factory floor.

0104

Apparel Craft

  • Apparel Production
  • Pattern Making
  • Garment Construction
  • 2D CAD
0204

Manufacturing Ops

  • Lean Management
  • Production Planning & Control
  • Apparel Quality Management
  • Machine Maintenance
0304

Intelligence

  • Data Analytics
  • R Studio
  • IoT Applications
  • Consumer Behavior Analysis
0405

Strategy & Soft

  • Marketing & Retail Planning
  • Problem Solving
  • Critical Thinking
  • Communication
  • Adaptability
Problem SolvingTime ManagementAttention to DetailWritten CommunicationActive ListeningCritical AnalysisAdaptableMS Excel · Word · PPTRetail EntrepreneurshipProblem SolvingTime ManagementAttention to DetailWritten CommunicationActive ListeningCritical AnalysisAdaptableMS Excel · Word · PPTRetail Entrepreneurship
[03]Case Studies

Seven projects.
One thesis: smarter, faster, kinder factories.

AGraduation · Smart Manufacturing

AI-Based Sewing Defect Prediction

A predictive intelligence layer for industrial sewing — built on ESP32 telemetry, Hall-effect and flex sensors, and a parameter-deviation model that calls defects before they happen.

  • Identified thread tension, machine speed and needle–hook ratio as the dominant defect drivers.
  • Instrumented a JUKI line with Hall, flex, temperature and humidity sensors streaming through ESP32.
  • Ran a Design-of-Experiments matrix across tension × speed × angle to train the classifier.
  • Real-time deviation engine flags stitches as normal, skipped, loose, broken or puckered.
ESP32Hall SensorFlex SensorDOEPythonReal-time Telemetry
Sensor-instrumented JUKI rig
Sensor-instrumented JUKI rig
Parameter relationship model
Parameter relationship model
DOE & data analysis flow
DOE & data analysis flow
3s
Pre-defect alert
7
Defect classes
JUKI
Industrial setup
BInternship · Predictive Ops

Downtime Prediction in Spreading

A Random Forest model trained on shop-floor variables to forecast spreading-line downtime — turning chaos on the cutting floor into a planning input.

  • Mapped Man / Machine / Material / Method causes via a fishbone root-cause analysis.
  • Cleaned and labelled live spreading data across multiple fabric and lay configurations.
  • Random Forest predicted downtime occurrence, timing and duration with operational accuracy.
Random ForestPythonRCAIndustrial Engineering
RCA across man, machine, material, method
RCA across man, machine, material, method
Downtime reduction breakdown
Downtime reduction breakdown
-50s
Loading time
↓
Re-laying minimized
↑
Productivity
CPredictive Maintenance

AI Machine Defect Clustering

An unsupervised maintenance model that clusters production machines by failure signature — pinpointing the exact machine where the next defect will land.

  • Built in R Studio using PCA-projected clusters across machine performance variables.
  • Categorised machines into immediate, scheduled and stable maintenance bands.
  • Hands recruiters of maintenance crews a ranked list, not a hunch.
R StudioClusteringPCAPredictive Maintenance
Cluster projection — R Studio
Cluster projection — R Studio
85%
Model accuracy
3
Maintenance tiers
PCA
Dimensionality
DOperational Analytics

Spreading Floor Downtime Reasoning

A second-pass model that doesn't just predict when downtime will hit — it predicts why, mapping causal reasons against machine and lay parameters.

  • Built parameter-aware reasoning across blade, software, ply, fabric and operator events.
  • Surfaces top downtime drivers per shift as a continuous decision-support feed.
R StudioClassificationCausal Reasoning
Reason × minutes — R Studio
Reason × minutes — R Studio
80%
Accuracy
12+
Cause classes
Live
Insights
ELean · 5S

Lean 5S Implementation

A merchandising-floor 5S deployment that turned sample chaos into a standardized library — measurable in the speed at which a sample can be pulled, sorted or discarded.

  • Audited received samples, sorted by relevance, set in standardized cartons.
  • Drafted a labelling and storage protocol the team could maintain independently.
  • Employees reported immediate ease in retrieval and reduced sample loss.
5SWorkflow DesignStandardization
5S — before / after on the merchandising floor
5S — before / after on the merchandising floor
Before
Chaos
After
Standard
↑
Throughput
FIoT · Textile Tech

Smart Drying Rack for Artisans

A sensor-instrumented drying rack designed for silk and other delicate fibers — alerts artisans when humidity, temperature and moisture cross the spoil-line.

  • Temperature and humidity sensors monitor microclimate in real time.
  • Moisture detection signals the extract-now moment for silk fibers.
  • Designed as a craft-friendly form factor — wood housing, embedded electronics.
IoTTemperature SensorHumidity SensorArtisan Tech
Prototype — silk-drying microclimate rack
Prototype — silk-drying microclimate rack
3
Live sensors
Silk
Optimised for
Alert
Auto extract
GWeb · Creative Concept

PromptMe — Author Platform

A self-developed brand and web product built in Adobe Dreamweaver, connecting authors and audiences directly for unfiltered idea exchange.

  • End-to-end brand, copy and visual identity for an original product concept.
  • Hand-coded in Dreamweaver — full responsive flow, original imagery.
  • Live on the web as a portfolio of fashion-tech-adjacent creative work.
Adobe DreamweaverBrand DesignWeb
promptme.netlify.app→
PromptMe — landing surface
PromptMe — landing surface
Live
Deployed
1
Solo build
100%
Original IP
[04]Atelier

From sketch to stitched.

A spread of pattern-engineered garments — each one built from the inside out, with techpacks, BOMs and operation breakdowns documented to manufacturing standard.

Look 01

Formal Trouser

Drafted and constructed a tailored formal trouser with front pleats — clean drape, considered seam architecture.

Trouser — Front / Side
Trouser — Front / Side
Look 02

Puff-Sleeve Dress

Pattern-engineered a dress with statement puff sleeves and a sculpted neckline; full BOM developed at Rs 812 total cost.

Dress — Front / Back
Dress — Front / Back
Skirt specification table
Skirt specification table
Look 03

Cargo Pants — Tech Pack

Twelve-pocket cargo silhouette taken from sketch to full techpack — flat sketches, BOM, operation breakdown and a QA matrix.

View techpack →
Cargo — Front / Back
Cargo — Front / Back
Flat sketches · BOM · QA
Flat sketches · BOM · QA
Look 04

Sleepsuit · Product Analysis

Analysed a knit baby sleepsuit and recreated it end-to-end — vendor-sourced fabrics and trims, full techpack with OB, packing and matched manufacturing cost.

Inspiration → Product developed
Inspiration → Product developed
[05]Technical Expertise

The full manufacturing stack.

01

Techpack Development

Full-spec techpacks with measurements, construction notes, vendor BOM and trim cards.

02

Bill of Materials

Manufacturing-accurate BOMs costed to the trim, fabric and consumable.

03

Quality Assessment

Quality assessment sheets and AQL-aligned inspection matrices.

04

Pattern Development

Drafted patterns from block to graded, in 2D CAD and on the table.

05

Manufacturing Analysis

Bottleneck mapping, cycle-time studies and line-balance for apparel lines.

06

Apparel Testing

Physical and dimensional testing protocols on fabric and finished garment.

07

Operation Breakdown

OB sheets that drive line layout, SAM and operator deployment.

08

Construction Process

Stitch class, seam type and machine selection for each construction detail.

[06]Manifesto

A working philosophy.

I

Fashion + Technology

"Garments are software now. Threads, sensors, telemetry — all one system."

II

Smart Manufacturing

"Factories don't need to be louder. They need to be smarter."

III

Data-Driven Creativity

"The most beautiful pattern is the one that survives the production line."

IV

Product Innovation

"Every techpack is a thesis. Every BOM is a budget for an idea."

V

Apparel Intelligence

"Predict the defect. Predict the downtime. Then design around it."

VI

Future of Fashion Systems

"Sustainable, predictable, human. In that order."

[07]Contact

Let's create smarter fashion systems together.

Email
ukrithika24@gmail.com
→
Phone
+91 9003365789
→
Open to
Apparel ManufacturingFashion-Tech RolesSmart Factory ProgramsProduct DevelopmentR&D · Innovation Labs