PRODUCT MANAGEMENT TRAINEE @ RVR PROJECTS PVT LTD · MICROSOFT CERTIFIED DP-600 & PL-300
I am Sasank Mangamuri — a Product Management Trainee at RVR Projects Pvt Ltd, dual Microsoft Certified in Microsoft Fabric (DP-600) and Power BI Data Analytics (PL-300), with hands-on expertise engineering production Laravel systems and publishing deep learning computer vision research in IEEE Xplore.
At RVR Projects Pvt Ltd, I drive technical product lifecycle workflows, operational analytics, project telemetry, and cross-functional coordination. Concurrently, I design OneLake medallion lakehouses, streaming PySpark ingestion pipelines, and Direct Lake semantic models capable of querying 50M+ rows with sub-1.2 second latency.
Beyond analytics, I engineer and deploy complete full-stack web applications with Laravel, PHP, MySQL, and Nginx on DigitalOcean cloud droplets, and co-authored an IEEE-published computer vision paper augmenting ConvNeXt with Coordinate Attention for chest X-ray diagnosis.
Leading technical product management workflows, operational analytics telemetry, infrastructure project tracking, and cross-functional coordination.
Engineered modular ERP and CRM systems in Laravel with role-based access control (RBAC), Razorpay automated payment workflows, and DigitalOcean droplet cloud infrastructure.
Managed digital asset telemetry, tracking platform uptime, user engagement KPIs, and generating operational performance insights.
Spearheaded operational data modeling, building automated Power BI dashboards and Python ETL scripts for product support telemetry.
Investigated convolutional architectures and attention mechanisms for multi-class medical diagnostic imaging, resulting in an IEEE-published paper.
End-to-end lakehouse engineering on Microsoft OneLake. Specializing in streaming PySpark ingestion, Delta Parquet partitioning, star-schema semantic modeling, advanced DAX calculation groups, and Direct Lake queries executing across 50M+ rows in under 1.2s without import refresh bottlenecks.
Microsoft Fabric (DP-600)
Power BI (PL-300)
PySpark
OneLake Architecture
Direct Lake
Delta Parquet
SQL & DAX
Designing and training attention-augmented neural networks with PyTorch. Authored peer-reviewed research on 1D Coordinate Attention with ConvNeXt backbones.
PyTorch
TensorFlow
ConvNeXt
Coordinate Attention
CLAHE
Architecting robust, scalable web platforms in Laravel and Python. Complete implementation of RBAC, payment gateway webhooks (Razorpay), automated invoicing, REST APIs, and high-performance database indexing.
Laravel
PHP 8+
Python
MySQL
JavaScript
Razorpay APIs
Deploying production workloads to DigitalOcean droplets with Nginx reverse proxies, Docker containerization, Let's Encrypt automated SSL renewals, and Google Analytics 4 telemetry tracking.
DigitalOcean
Docker
Nginx
GitHub Actions
Google Analytics 4
Figma
Microsoft Fabric Power BI (PL-300) Python PySpark Laravel MySQL PyTorch DigitalOcean Docker Figma Microsoft Fabric Power BI (PL-300) Python PySpark Laravel MySQL PyTorch DigitalOcean
# OneLake Raw Ingestion Pipeline (Bronze)
from pyspark.sql import functions as F
bronze_df = (spark.readStream
.format("cloudFiles")
.option("cloudFiles.format", "json")
.option("cloudFiles.schemaLocation", "abfss://onelake@fabric/schemas/bronze")
.load("abfss://onelake@fabric/landing/")
.withColumn("ingestion_timestamp", F.current_timestamp())
.withColumn("source_system", F.lit("Shanix_ERP")))
(bronze_df.writeStream
.format("delta")
.outputMode("append")
.option("checkpointLocation", "abfss://onelake@fabric/checkpoints/bronze")
.toTable("lakehouse.bronze_events"))
High-scale multi-layer OneLake medallion lakehouse with PySpark ETL ingestion, Delta Parquet partitioning, and Direct Lake semantic models querying 50M+ rows in under 1.2s without import refresh bottlenecks.
Modular enterprise CRM and ERP platform built in Laravel featuring granular RBAC, Razorpay automated payment webhook handling, dynamic invoice generation, and DigitalOcean droplet deployment.
Peer-reviewed chest X-ray classifier combining ConvNeXt-Nano with 1D Coordinate Attention pooling for 4-class pulmonary diagnosis on the COVID-19 Radiography Dataset. Published in IEEE ICIMA 2026.
Multi-role learning platform with student & instructor portals, automated attendance tracking, quiz grading engine, and administrative engagement analytics dashboards.
End-to-end shopping platform with product catalogs, shopping cart state management, checkout with Razorpay gateway, and deployment on DigitalOcean with Nginx SSL.
Verified expertise in OneLake architecture, Delta Parquet partitioning, PySpark data ingestion, and Direct Lake semantic model performance optimization.
Power Query ETL, star schema design, advanced DAX measures, calculation groups, row-level security (RLS), and enterprise report distribution.
Event tracking, conversion funnels, audience segmentation, user cohorts, and attribution modeling (Skillshop ID 186505260).
Pandas, NumPy, data visualization, REST APIs, and foundational machine learning workflows in Python.
Data cleaning, data modeling, dashboard visualization, and executive stakeholder presentation strategy.
Customer acquisition journeys, conversion optimization, measurement frameworks, and multi-channel campaign analytics.
Accurate automated identification of pulmonary conditions from chest radiography requires fine-grained spatial features without heavy self-attention computational cost. We augment ConvNeXt with 1D Coordinate Attention — applying 1D pooling across horizontal and vertical axes — and benchmark on the COVID-19 Radiography Dataset across normal, viral pneumonia, lung opacity, and COVID-19 classes with CLAHE contrast enhancement.
Toggle preprocessing stages and directional attention projections evaluated on 21,165 chest radiography scans.
| Architecture Backbone | Attention | Accuracy | Precision | Latency |
|---|---|---|---|---|
| ConvNeXt-Nano (Ours) | 1D Coord Attn | 98.4% | 98.1% | 14.2ms |
| Vision Transformer (ViT-B/16) | Self-Attention | 95.2% | 94.8% | 41.5ms |
| DenseNet-121 | None | 94.6% | 94.1% | 29.1ms |
| ResNet-50 (Baseline) | None | 93.1% | 92.5% | 22.8ms |
Product Management Trainee at RVR Projects Pvt Ltd. Reach out for technical architecture discussions, research collaborations, or speaking engagements.