Srimanth | AI Acceleration Architect @ AWS

srimanth — zsh — 80×24

Featured Projects

--sort=impact

$ ls ./projects --featured --limit=4

nba-correlation-pipeline.json

"name": "NBA Correlation Analysis Pipeline",

"tagline": "End-to-end pipeline from web scraping to context-stratified correlation and DVP analysis across 27,990 player-game records",

"stack": ["Python", "pandas", "scipy", "BeautifulSoup", "ML"],

"records processed": 27,990,

"correlations calculated": 89,181

"status": "live"

$ view-project →

"name": "LingoPT",

"tagline": "AI-powered English speaking coach built as a Christmas gift for my mom",

"stack": ["AWS", "GenAI", "React", "Serverless"],

"aiModel": "Claude 3.5 Sonnet",

"awsServices": "5 (Bedrock, Polly, Transcribe, S3, Lambda)"

"status": "live"

$ view-project →

nfl-prediction-system.json

"name": "NFL Matchup Prediction System",

"tagline": "Percentile-based NFL matchup predictions using EPA metrics and coverage scheme analysis",

"stack": ["Python", "Statistical Modeling", "EPA Metrics", "pandas", "numpy"],

"methodology": "Percentile-based",

"metrics tracked": "EPA, success rate, directional tendencies"

"status": "live"

$ view-project →

nba-ar-stats.json

"name": "NBA AR Stats Viewer",

"tagline": "iOS AR app that identifies NBA players and overlays live stats in augmented reality",

"stack": ["Swift", "ARKit", "RealityKit", "AWS Rekognition", "iOS", "Computer Vision", "NBA"],

"platform": "iOS",

"ar framework": "ARKit + RealityKit"

"status": "complete"

$ view-project →

$ ls ./projects --all→

Current Role

--latest

aws-partner-sa.sh

$ role

Partner Solutions Architect/AI Acceleration Architect @ Amazon Web Services

$ period

August 2024 — Present

AWS CloudGenerative AI - AI/MLMCP ProtocolClaude CodePython

$ view-role →

$ cat ./built-with-ai.md

This entire site was pair-programmed with AI

Every component, animation, and deployment pipeline was built using AI-assisted development. See the prompts, the process, and what AI-augmented development actually looks like.

$ view --built-with-ai→