Engineering Portfolio
I design and deploy complex
AI systems at scale
Here's proof that I build real-world systems with impact, performance, and architecture in mind.
VaidyaOS
Advanced AI healthcare platform for intelligent, real-time, multilingual medical assistance using on-device edge AI and offline capabilities via lightweight GGUF models.
Offline-first architecture with Edge AI inference & Llama.cpp
CareerLens
Built an AI-powered career platform that analyzes resumes, identifies skill gaps, and generates personalized career paths. Designed as a microservices system and deployed for real-world usage.
Event-driven microservices architecture with parallel embedding queues
AgriSence
Developed a real-time AI farming system for crop disease detection and advisory. Integrated ML models with satellite data and multilingual support, deployed on GCP for scalable access.
Serverless ML inference pipeline with real-time edge sync
OmniSence
Engineered a multimodal creative storytelling engine processing text to audio and video. Combined LLMs, Imagen, and Cloud TTS for real-time narrative generation.
Real-time streaming pipeline with async media generation
TaskForze
Created a multi-agent productivity system that coordinates 5 specialized AI sub-agents to manage workflows, with automated escalation via WhatsApp and AI voice calls for missed deadlines.
Multi-agent orchestration system with vector retrieval
HealthMesh v2.0
Built a HIPAA-ready clinical AI platform combining 6 specialized healthcare agents with a prescription bridge API to link clinical insights to hyperlocal medicine delivery.
FHIR R4 compliant microservices with secure data enclave
CyberShield AI
Developed and deployed multiple systems including a 5-layer SME cybersecurity suite (scoring 78/100) and an offline C-based mesh communication network for emergency triage.
What these projects demonstrate
Production AI Systems
Ability to go beyond prototypes to build, deploy, and scale real-world generative AI and multimodal ML pipelines.
Scalable Cloud Infrastructures
Designing complex microservices and backend architectures using GCP, Azure, serverless deployments, and vector databases.
Full-Stack Autonomy
Working across the entire stack—from frontend UI/UX in Next.js to backend APIs in FastAPI, all the way to secure database schemas.
Practical Real-World Impact
Solving actual problems like clinical orchestration, offline mesh emergency networks, and crop disease mapping through system design.