Omkar Chebale

@Chebaleomkar · User

GitHub profile ↗ · Compare

Inference | AI/ML Engineer

7 followers86 repositories

Repositories

Chebaleomkar/ZeroHost-vLLM

Production-style LLM inference on free compute: vLLM + Qwen2.5-7B on a Kaggle T4, OpenAI-compatible API, tool calling, full tracing and benchmarks.

★ 0HTMLForks 1

Chebaleomkar/ZeroHost-vLLM-v2

Single GPU vs data parallel vs tensor parallel on Kaggle's free 2x T4: vLLM + Qwen2.5-7B, load-balancing proxy, in-session benchmarks, OpenAI-compatible API.

★ 0PythonForks 0

Chebaleomkar/OpenBot

Open-source AI coworkers that each get a computer of their own: a browser, files and tools, with every action decided before it happens and recorded after. Bring any AG-UI agent.

★ 0Forks 0

Chebaleomkar/company-brain

Open-sourcing our company brain - A teammate in your Slack that remembers everything your team says, and can go do the work.

★ 0Forks 0

Chebaleomkar/threadcut

A mini Subconscious Cache: pruning an agent's context with prefix + suffix KV reuse, measured with a real coding agent on a free Kaggle T4

★ 0PythonForks 0

Chebaleomkar/hive

Outcome driven agent development framework that evolves

★ 0PythonForks 0

Chebaleomkar/edu-agent-pipeline

An agent-based AI application that generates, reviews, and refines educational content through a transparent multi-step workflow. It uses specialized generator and reviewer agents, structured JSON outputs, and a simple UI to demonstrate practical AI agent pipelines for real-world use cases.

★ 0PythonForks 0