DEV Community: upendra manike

# 🚀 Running Claude Code Locally with Ollama (No Token Cost)

upendra manike — Sat, 11 Apr 2026 19:29:28 +0000

Until recently, using Claude for coding workflows meant relying on paid API usage.

Now, there’s a powerful workaround:

👉 You can run Claude Code against a local Ollama endpoint, using open-source models like qwen2.5:3b.

This enables a fully local AI coding assistant — no per-token billing, and full control over your environment.

⚙️ Setup Guide

1. Install Ollama

brew install ollama

2. Pull a Coding Model

ollama pull qwen2.5:3b

3. Install Claude Code

npm install -g @anthropic-ai/claude-code

4. Configure Local Endpoint

export ANTHROPIC_AUTH_TOKEN=ollama
export ANTHROPIC_BASE_URL=http://localhost:11434

5. Run Claude Code Locally

claude --model qwen2.5:3b

🧠 What This Actually Does

Instead of sending requests to Anthropic’s servers, Claude Code:

Calls a local API (Ollama)
Uses an open-source LLM
Executes agentic workflows on your machine

✨ Benefits

No API cost → completely free usage
Privacy-first → your code never leaves your system
Flexible models → switch between different open-source LLMs
Offline capability → works without internet

⚠️ Limitations

Let’s be honest:

Not equivalent to Claude Sonnet/Opus quality
Smaller models struggle with complex reasoning
Performance depends on your hardware

For example:

3B models → fast but limited
7B–13B → balanced
30B+ → powerful but slow on laptops

💡 When to Use This

Best use cases:

Local development assistant
Code autocomplete / small tasks
Privacy-sensitive projects
Cost-sensitive workflows

🚀 Final Thoughts

This setup represents a shift toward:

Local-first AI development

While cloud models still lead in performance, local setups are becoming increasingly practical for everyday workflows.

And for developers, this means:

👉 More control
👉 Lower cost
👉 Faster experimentation

⭐ If you're building with local AI agents, I’d love to hear your setup.

🚀 I Built an Open Source AI Code Assistant (OpenWorkbench-AI)

upendra manike — Sat, 11 Apr 2026 19:19:08 +0000

AI is rapidly changing how we write code.

From autocomplete tools to full-blown AI agents, developers are no longer just writing code — we are orchestrating intelligence.

So I decided to build something myself 👇

🔗 https://github.com/upendra-manike/OpenWorkbench-AI/tree/main/ai-code-assistant

💡 What is OpenWorkbench AI Code Assistant?

OpenWorkbench AI is an AI-powered coding assistant designed to:

Generate code from natural language
Help debug and refactor existing code
Understand multi-file projects
Work with both local models and APIs

The goal is simple:

Give developers a Claude/Copilot-like experience — but with flexibility and control.

⚙️ Why I Built This

Most AI coding tools today are:

Closed-source
Expensive (token-based pricing)
Limited in customization

As developers, we need:

Control over models (local + cloud)
Custom workflows
Extensibility

This project is my step toward an open AI engineering ecosystem.

🧠 Key Features

🔹 AI Code Generation

Write prompts like:

“Create a REST API in Spring Boot”

Get structured, usable code instantly.

🔹 Debugging & Refactoring

Analyze existing code
Suggest improvements
Fix issues faster

🔹 Multi-Model Support

Use local models (Ollama)
Or plug in APIs like Claude/OpenAI

🔹 Developer-Centric Design

Built with:

FastAPI backend
Modular architecture
Easy extensibility

🏗️ Architecture Overview

User Prompt
   ↓
AI Agent Layer
   ↓
--------------------------------
| LLM (Local / API)            |
| Prompt Engine               |
| Code Processing Layer       |
--------------------------------
   ↓
Generated Code / Suggestions

🔥 What I Learned

Building this taught me:

AI is not just about models — it’s about system design
Prompt engineering is a real engineering skill
Latency + cost optimization matter in production
AI agents are the future of development workflows

🚀 What’s Next

Planned features:

Full project generation (multi-screen apps)
Integration with VS Code / Cursor
Autonomous dev agents
Testing + CI automation

🤝 Open for Contributions

If you’re interested in:

AI engineering
LLM applications
Developer tools

Feel free to contribute or share feedback 🙌

📌 Final Thought

We are moving from:

Writing code → Designing intelligent systems

And this is just the beginning.

⭐ If you like the project, give it a star!

I Built a Chrome Extension – Here’s What I Learned

upendra manike — Wed, 11 Mar 2026 19:39:07 +0000

I recently built a small Chrome extension and published it.

🔗 https://chrome-stats.com/d/phiikbecemoodpnmlngbcfnfjgpikkhb

Chrome extensions are essentially small applications that extend browser functionality, allowing developers to modify website behavior or add new capabilities directly inside the browser.

Why I Built It

As developers, we spend most of our time inside the browser. Creating lightweight tools that improve productivity can have a surprisingly large impact.

What I Learned

Building browser extensions teaches you several useful concepts:

working with browser APIs

handling background scripts and content scripts

managing permissions and security

building lightweight automation tools

Why Extensions Are Interesting

Unlike traditional applications, extensions run directly inside the user's browser, which makes them great for productivity tools, automation, and developer utilities.

If you're a developer and haven't tried building a Chrome extension yet — it's definitely worth exploring.

🚀 Built a One-Click UI Testing Chrome Extension to make web testing dramatically faster.

upendra manike — Tue, 10 Mar 2026 18:24:38 +0000

While working on multiple web projects, I noticed that writing UI tests is often time-consuming, repetitive, and difficult to maintain. Test engineers and developers spend hours writing selectors, configuring frameworks, and debugging fragile test scripts.

So I built a Chrome Plugin that generates UI tests with a single click.

What the extension does

• Record user actions directly from the browser
• Generate ready-to-run UI test scripts automatically
• Capture selectors and interactions without manual coding
• Export tests for automation frameworks
• Help developers and QA teams quickly validate UI workflows

Instead of manually writing test scripts, you can simply interact with the application and let the plugin generate the test logic for you.

Why I built this

UI testing should not slow down development.
The goal of this tool is to make test generation simple, fast, and accessible to every developer and QA engineer.

Try the extension here:
https://chromewebstore.google.com/detail/phiikbecemoodpnmlngbcfnfjgpikkhb?utm_source=item-share-cb

I’m currently refining the product and would love feedback from:

• QA engineers
• Developers building web applications
• Teams struggling with UI automation

If you're interested in trying it or collaborating, comment “UI TEST” or send me a message.

Testing #Automation #ChromeExtension #QA #SoftwareTesting #WebTesting #DeveloperTools #Startup #ProductBuild

🚀 Built a One-Click UI Testing Chrome Extension that records user actions and automatically generates ready-to-run UI test scripts. No manual selectors or fragile scripts—just interact with the app and export tests instantly. https://acesse.one/wXtLs

upendra manike — Tue, 10 Mar 2026 18:20:01 +0000

encurtador.dev