01 — HIGH CONCURRENCY & INFERENCE AT SCALE
Software
I engineer high-throughput backend architectures, low-latency AI inference pipelines, and structured telemetry databases. Allergy to latency, devotion to reliability.
THE SHIPPED CODEBASE
Featured software work
FETTLE · FOUNDING ENGINEER
Fettle: Healthcare Voice AI
Distributed voice agents for hospital customer experience and automated workflows. Orchestrated high-concurrency SIP trunking with LiveKit, built a cross-modal clinical recommender engine, and optimized real-time sub-second inference using PyTorch and TorchRec on Kubernetes. Owned the complete backend design and DevOps lifecycle.
Agentic AI
LiveKit
PyTorch
Kubernetes
FastAPI
Inference
SOLARMOBIL · DATA ENGINEERING
Solar Racing Digital Twin
A real-time telemetry and race-strategy engine for a solar EV. Fused real-time GIS elevation data, live weather APIs, and vehicle state dynamics using a stochastic receding-horizon model predictive control framework. Handled time-series data storage on InfluxDB and designed real-time observability dashboards with Grafana.
InfluxDB
Grafana
Data Engineering
Control Theory
Python
Observability
GUAQ AI · FOUNDING WORK
Teachafy Labs
Enterprise generative AI solutions. Built complex multi-agent workflows and high-performance retrieval-augmented generation (RAG) pipelines. Developed custom integrations linking vector databases directly to corporate ERP/CRM systems, enabling sub-second factual query answering.
RAG
Agentic AI
LangChain
Vector DB
ERP
DCPR AI · PIP PACKAGE
XX-TB Scale Document Processing
Distributed multi-TB document ingestion pipeline using GCS and Gemini. Cut manual processing queues by 92% and storage costs by 74%.
Kubernetes
GCS
Redis
Distributed System
MIT, MANIPAL · TICKETING & SCALE
Dev Team Lead
Led the developer division for national-level college fests. Designed and deployed highly concurrent ticketing, registration, and payment gateway platforms. Maintained 99.9% uptime and handled tens of thousands of simultaneous requests under heavy peak loads.
System Design
Ticketing
Node.js
Chakra UI
Scale
THE TECHNICAL STACK
What I write & deploy
Languages
ML & AI
Web & Full-Stack
Cloud & DevOps
THE CORE FOCUS
What I'm good at
Agentic AI & Inference
Building autonomous LLM workflows and optimized inference pipelines using LiveKit, PyTorch, and distributed K8s clusters.
Data Engineering
Implementing real-time signal acquisition and processing pipelines for bio-signals and telemetry.
Distributed Systems
Developing scalable backends with FastAPI, Kubernetes, and cloud-native observability stacks.