CHIRUDEVA REDDY // RESEARCH SCHOLAR & SDE
0%
CHIRUDEVA REDDY // AI ENGINEER, WITH A RESEARCH INTEREST IN CV

Building perceptual intelligence for complex physical systems.

3rd-year Computer Science undergraduate at BITS Pilani Dubai with a Minor in Data Science, actively contributing to 3D computer vision, big data ML pipelines, high-throughput NLP engines, and scalable system design.

EXPLORE PROJECTS ↗ RESEARCH FOCUS
DUBAI, U.A.E  •  S26 ACTIVE
ACADEMIC DISTINCTION (YTD)
9.13
CGPA / 10.0 • BITS Pilani Dubai (Conferred 09/2027)
PRODUCTION ERROR REDUCTION
42%
API endpoint debugging & low-latency optimization at Nablix Labs
BIG DATA RECORDS ANALYZED
1M+
Chicago dataset crash risk modeling with XGBoost
NN TRAINING ERROR CUT
98%
Continuous depression risk model under Dr. Shanookha Ali
RESEARCH & ENGINEERING PHILOSOPHY

Extracting dense 3D representations from sparse privacy-preserving inputs.

My academic research focuses on the intersection of Computer Vision, Deformable 3D Representations, and Privacy-Preserving Perception. Under the supervision of Dr. Shivang Agarwal, I am authoring a paper proposal for IJCAI on estimating parametric 3D human body meshes directly from 2D binary silhouettes without storing raw optical imagery.

In industry, I served as AI Engineer Intern at Nablix Labs (curating API endpoints and reducing error rates by 42%), Fullstack Developer Intern at Jovens Technologies FZ-LLC (architecting fullstack ML apps with 28% efficiency boost and NLP pipelines reducing manual time by 12%), and SDE Intern at Domdog (building Playwright automation, malicious content scanner integration across 8 client platforms, and boosting SEO traffic by 11%).

CORE COMPETENCIES & SKILLS

Technical Matrix & Stack

CORE LANGUAGES

Python & SQL

Proficient in Python, SQL data querying; familiar with Java, C, C++, and high-performance algorithms.

PERCEPTION & 3D MESH

Computer Vision & SAM

OpenCV, YOLOv8, Segment Anything Model (SAM, SAM-2), MediaPipe pose estimation, and 3D mesh extraction.

CORE MACHINE LEARNING

PyTorch & Scikit-Learn

Custom neural architectures, Transfer Learning, Feature Engineering, Model Evaluation, and CUDA acceleration.

NLP & TRANSFORMERS

NLP & Text Clustering

Sentiment extraction, text preprocessing, transformer embeddings, BERT fine-tuning, and semantic clustering.

BIG DATA ANALYTICS

Data Science & Pipelines

Pandas, NumPy, Matplotlib, exploratory data analysis, tabular data preprocessing, and risk visualization.

SYSTEM DESIGN & APIS

Fullstack & Backend

FastAPI microservices, full-stack ML apps, POST response validation pipelines, and JSON serialization.

CLOUD INFRASTRUCTURE

AWS S3 & Azure

AWS S3 storage pipelines, Microsoft Azure cloud services, Docker containerization, and data ingestion.

AUTOMATED VERIFICATION

Playwright & Security

Chromium automation, consumer simulation, malicious content scanners, test coverage expansion, and CI/CD.

BI & SCIENTIFIC TOOLING

PowerBI, Tableau & Git

PowerBI, Tableau dashboards, LaTeX scientific documentation, Git/GitHub version control, Excel, and SEO/GEO.

ENGINEERING PORTFOLIO

Featured Projects & Architecture

04 / 04
CV // IJCAI PROPOSAL (09/2025 – PRESENT)

Privacy-Preserving Body Composition Estimation

Creating a pipeline to estimate human body composition from silhouettes under privacy constraints using transfer learning, hyperparameter tuning, and 3-D reconstruction under supervision of Dr. Shivang Agarwal.

REPOSITORY ↗
mesh_reconstruct.py PYTORCH // SMPL // TRANSFER LEARNING 
# 3D Silhouette Reconstruction
def estimate_body_composition(silhouette_2d):
    params = smpl_estimator.predict(silhouette_2d)
    mesh_3d = smpl_model.get_vertices(params)
    return optimize_privacy_constraints(mesh_3d)
NLP // INTELLIGENCE (01/2026 – PRESENT)

NLP Insight Engine for Customer Reviews

Working on creating an NLP pipeline to extract sentiment and key themes from large-scale customer reviews using text preprocessing, transformer embeddings, and clustering algorithms.

REPOSITORY ↗
review_nlp_engine.py TRANSFORMERS // BERT // CLUSTERING 
# Large-Scale Customer Review NLP
class ReviewInsightEngine:
    def extract_themes_and_sentiment(self, reviews):
        clean_text = self.preprocess_text(reviews)
        vectors = bert_model.encode(clean_text)
        return kmeans_cluster(vectors, k=12)
DATA SCIENCE // 1M+ RECORDS (09/2025 – 01/2026)

Road Accident Severity Prediction

Built a collaborative 4-member project on crash severity prediction for the Chicago Crash Dataset consisting of over 1,000,000 records. Contributed data-preprocessing and interactive risk visualization.

REPOSITORY ↗
crash_severity.py XGBOOST // CHICAGO 1M+ DATASET 
# 1M+ Records Geospatial Preprocessing
df = pd.read_csv("chicago_crash_1M.csv")
X_features = engineer_spatial_and_temporal(df)
model = XGBClassifier(n_estimators=300)
model.fit(X_features, y_severity)
AI // FUZZY & ANN (02/2024 – 05/2024)

Predicting Student Depression Risk

Worked under Dr. Shanookha Ali to build a system using fuzzy logic and artificial neural networks in order to tune and predict continuous depression risk of students from lifestyle and academic factors, achieving 98% training error reduction.

REPOSITORY ↗
depression_risk_ann.py FUZZY LOGIC // NEURAL NETWORKS 
# Fuzzy Logic & Neural Network Optimization
class ContinuousDepressionRiskPredictor:
    def evaluate_lifestyle_factors(self, lifestyle_vector):
        fuzzy_scores = self.fuzzy_inference(lifestyle_vector)
        return self.ann_model.predict(fuzzy_scores)
PROFESSIONAL EXPERIENCE

Industry Experience & Track Record

AI Engineer Intern

Nablix Labs • Dubai, U.A.E (Hybrid)
06/2026 – 09/2026
  • Curated and optimized API endpoints along with a strong focus in debugging issues and reducing the error rate by 42% before production.
  • Engineered low-latency response serialization and automated input telemetry handling.
PyTorch FastAPI API Debugging Model Optimization Python

Fullstack Developer Intern

Jovens Technologies FZ-LLC • Dubai, U.A.E
02/2026 – 08/2026
  • Architected and deployed a scalable full-stack web application, integrating Machine Learning (ML) and Artificial Intelligence (AI) algorithms to optimize core platform features, resulting in a 28% increase in processing efficiency.
  • Engineered robust Natural Language Processing (NLP) pipelines for automated text analysis and unstructured data extraction, reducing manual processing time by 12% and improving overall system accuracy by 15%.
Fullstack Machine Learning NLP Pipelines Text Extraction System Architecture

Software Development Engineer Intern

Domdog • Hyderabad, India (Hybrid)
06/2025 – 09/2025
  • Created a backend response collection pipeline to validate POST requests, serialize JSON, and store data.
  • Developed Python and Playwright automation in Chromium to simulate consumer behavior leading to a 10% increase in test case scenarios.
  • Integrated company-developed malicious content scanner with 8 client systems in 3 months.
  • Optimized database querying and data entry, reducing turnaround time by approximately 3 days.
  • Curated technical documentation for client integrations and applied Search/Generative Engine Optimization (GEO/SEO), increasing website traffic by 11%.
Playwright Python Security Audit JSON Pipelines SEO / GEO Database Querying