---
title: "Cost-Aware AI Gateway | Vivek Singh"
url: https://vivekbuilds.com/work/router/
description: "Model selection that balances task fit, cost and latency, with caching and provider failover."
---

# Cost-Aware AI Gateway

An inference gateway that selects a model for the task, optimizes request handling and manages cost across AI providers.

## Architecture
- Task-aware routing across OpenAI, Claude and Gemini
- Quality, cost and latency tradeoffs backed by evaluation
- Caching, circuit breakers and cross-provider failover
- Token/cost monitoring, tracing and evaluation





Descriptive project name. Portfolio interfaces are illustrative.