SYST.DEV
sensors
terminal v2.4.0-stable_arch

ENGINEERING INTELLIGENT BACKENDS.

Specializing in distributed AI infrastructure and high-throughput systems. Architecting the next generation of neural-integrated services with Rust, Go, and Kubernetes.

profile.json — aviral_tyagi
{
"identity": "Aviral Tyagi",
"role": "Backend Engineer",
"focus": [
"Distributed Systems",
"LLM Infrastructure",
"Edge Computing"
],
"status": {
"availability": true,
"location": "India"
},
"active_threads": 128
}

Work Experience

EXPERIENCE_LOG.BIN

[Jul. 2023 - Jun. 2024]

Backend Engineer

Rupifi, Bangalore

Developed and maintained scalable backend services using Spring Boot & PostgreSQL enabling efficient frontend integration, microservices communication, and third-party platform connectivity.

Integrated Exotel services to implement device binding workflows, increasing mobile app user device binding adoption from 20% to 70%.

Optimized API performance through query tuning, external API call improvements, and cursor-based pagination, reducing response times to under 100 ms and eliminating timeout issues.

Improved API response times by over 200% through database query optimization and efficient handling of external service dependencies.

Automated daily reconciliation of database mismatch records by developing REST APIs, building Python automation scripts, and deploying scheduled workflows on AWS Lambda.

[Jul. 2022 - Jun. 2023]

Backend Engineer Intern

Rupifi, Bangalore

As part of credit line team developed multiple REST APIs using Springboot and PostgreSQL.

Reduced collections team’s work from hours to a minute by developing tool to automate their work using AppScript in google sheet.

Wrote python script for CRON job and deployed it using AWS Lambda.

Deployed service using AWS Codebuild and used AWS Cloudwatch & OpenSearch for logs.

[Jul. 2022 - Jun. 2023]

ML Research Intern

Maritime Research Center, Pune

Implemented and simulated the Wittekind Underwater Noise Model to study underwater acoustic signatures of maritime vessels, generating high-quality training data that improved the effectiveness of downstream regression models for noise level prediction.

Applied supervised machine learning regression algorithms (random forest, gradient boosting tree, and decision tree).

Achieved accuracy of 87% considering predicted noise value difference from the original is 5db or less.

Academic History

EDUCATION_REGISTRY.LOG

[2025 - Present]

MTech. Computer Science and Engineering

IIIT Hyderabad

[2018 - 2023]

MSc. Physics & B.E Electrical & Electronics

BITS Pilani, Pilani

Active Deployments

PROJECT_REGISTRY.BIN

PROJ_01

DISTRIBUTED_MODEL_TRAINING

A cluster orchestration tool for parallelizing large-scale transformer training across heterogeneous cloud providers.

PROJ_02

AUTONOMOUS_API_GATEWAY

A zero-trust gateway with self-healing routing and real-time anomaly detection powered by eBPF.

PROJ_03

HYPER_GRAPH_DB

Ultra-low latency graph engine optimized for relationship heavy AI workloads and real-time knowledge graphs.