Projects

Production-grade AI systems, multi-agent pipelines, RAG architectures, and ML projects.

AI Agent for Engineering Design Simulation & Analysis
AI/ML

Agentic AI system that autonomously parses complex simulation logs, identifies anomalies, and summarizes key performance metrics using CrewAI and LangChain.

Python
CrewAI
LangChain
PyTorch
FAISS
+2

Reduced manual review time by 75%

Multi-Agent System for Technology & Market Evaluation
Featured
AI/ML

Multi-agent platform where autonomous agents collaborate to perform real-time evaluation of business and technology proposals — Flask backend, React frontend.

CrewAI
GPT-4o
Flask
React
Python
+1

Full-stack AI-powered decision support

Agentic City Data Automation System
Featured
AI/ML

Agentic pipeline that ingests city email notifications, scrapes city websites, extracts meeting details, and generates structured calendars, agendas, and summaries.

Python
FastAPI
LLMs
Web Scraping
MongoDB
+2

1M+ records/week · ~30% accuracy improvement

KG-RAG Hybrid Retrieval System
Featured
RAG/KG

Hybrid retrieval system combining Neo4j knowledge graphs with vector databases (Weaviate, FAISS) for multi-hop reasoning and context-aware search.

Neo4j
Weaviate
FAISS
RAG
Python
+2

Improved retrieval relevance by ~30%

IoT Pet Health Monitoring — ML Pipeline
AI/ML

ML pipeline for an AI-driven hardware pet health product, integrating on-device firmware with cloud backend and processing high-frequency IoT telemetry.

Python
PyTorch
IoT
FastAPI
AWS
+2

Real-time pet health anomaly detection

City Meeting Transcript Scraper
AI/ML

Web app to automatically scrape and process transcripts from city meeting videos, enabling downstream NLP and summarization pipelines.

Python
Web Scraping
FastAPI
NLP
PostgreSQL

Feeds 1M+ records/week pipeline

Business Data Pipeline & Reporting Automation
Data Science

Internal Python + Shell data pipelines to automate inventory tracking, sales reporting, and analytics dashboards — driving 30% YoY product growth.

Python
Shell
SQL
ETL
Data Pipelines
+2

30% YoY growth · 60% less manual reporting

Private / Internal
Local LLM & GPU Systems Lab
Featured
GPU Systems

A systems-focused project exploring how local LLMs behave under different memory, inference, and fine-tuning conditions — GPU memory allocation, quantization, context length, and benchmarking across models.

PyTorch
CUDA
Local LLMs
Quantization
QLoRA
+2

VRAM, tokens/sec & load-time benchmarks across models

GPU Memory & Transfer Lab
GPU Systems

Low-level experiments on GPU memory behavior, CPU–GPU transfer costs, and runtime performance — allocated vs reserved VRAM, pinned memory, async copies, streams, and CUDA event timing.

PyTorch
CUDA
GPU Memory
Pinned Memory
CUDA Streams
+1

A practical view of what happens below the model API

Local LLM Runtime Benchmark
GPU Systems

A benchmarking harness comparing local LLM inference across model sizes, quantization modes, and context lengths — tracking tokens/sec, peak VRAM, model load time, and output quality.

Local LLMs
Benchmarking
Quantization
VRAM
Inference
+1

Disciplined benchmarking over subjective impressions

QLoRA Fine-Tuning Experiment
Deep Learning

A fine-tuning project adapting a local model with QLoRA while tracking VRAM, training loss, and before/after output quality on a CUDA & quantization instruction dataset.

QLoRA
LoRA
4-bit Quantization
Fine-Tuning
PyTorch
+1

Lower loss ≠ better answers — measured behavior change

GPU Workload Scheduler Simulation
GPU Systems

A systems simulation of how multiple workloads share a GPU — comparing FIFO, priority, round-robin, and deadline-aware scheduling against p50/p95/p99 latency, throughput, and deadline misses.

Scheduling
Simulation
Latency
Throughput
GPU
+1

Predictability & tradeoffs in shared compute

Camera-to-GPU Pipeline
Computer Vision

A computer-vision pipeline measuring latency from capture to GPU processing — frame capture, CPU preprocessing, CPU→GPU transfer, inference, and end-to-end frame latency / FPS.

Computer Vision
CUDA
Real-time
Latency
PyTorch
+1

Stage-by-stage latency in real-time perception

Quick Lube & Smog Website
Web Dev

A production website designed and developed end-to-end for a local automotive business — services, trust signals, and contact, from requirements through deployment and maintenance.

Web Development
Frontend
Design
Deployment
SEO

Live customer-facing site

Mass Communication Platform
Featured
Web Dev

A full-stack communication platform (FastAPI, Next.js, MongoDB) for scalable outreach — contact management, API integrations, and reliable message-delivery workflows.

FastAPI
Next.js
MongoDB
REST APIs
Full-Stack
+1

Scalable outreach & message delivery

AI Search & Knowledge Platform
Featured
RAG/KG

A document search and retrieval system for natural-language questions over large collections — combining vector search, metadata filtering, and LLM summarization behind backend APIs.

RAG
Vector Search
Embeddings
LLMs
FastAPI
+1

Natural-language search over large document sets

More experiments, research notebooks, and open-source work on GitHub.

View GitHub Profile