N150 N300 T3K P100 P150 P300C Galaxy 15 min Draft

Coding Assistant with Aider

Overview

Aider is an open-source AI pair programming tool that runs in your terminal. It reads your codebase, edits files, and auto-commits changes — and it speaks the OpenAI API, so you can point it directly at your Tenstorrent vLLM server.

Works on all hardware — N150/N300/T3K/P100/P300c/Galaxy. No ~/tt-metal required.

Prerequisites: Your vLLM server must be running before launching Aider. See vLLM Production to start it.

Step 1: Install Aider

pip install aider-chat

That's the only install. Aider is a self-contained Python package with no tt-metal dependencies. Install it in any environment — your system Python, a separate venv, or even pipx:

# Isolated install (keeps dependencies separate from tt-metal venv)
pipx install aider-chat

Step 2: Connect to Your Local LLM

Set two environment variables to point Aider at your running vLLM server:

export OPENAI_API_BASE=http://localhost:8000/v1
export OPENAI_API_KEY=fake       # vLLM doesn't check the key; any value works

Then launch Aider with the model you started in vLLM:

# If you started vLLM with Qwen3-0.6B (works on all hardware):
aider --model openai/Qwen3-0.6B

# If you started vLLM with Qwen3-8B or Llama 3.1 8B:
aider --model openai/Qwen3-8B
aider --model openai/Llama-3.1-8B-Instruct

The openai/ prefix tells Aider to use your custom API base instead of calling OpenAI's servers.

Step 3: Make It Permanent (Optional but Recommended)

Drop a .aider.conf.yml in your project root to skip the env vars and flags on every launch:

# .aider.conf.yml  (commit this to your repo!)
openai-api-base: http://localhost:8000/v1
openai-api-key: fake
model: openai/Qwen3-0.6B

Now in that project directory, just run:

aider

No flags, no env vars. Aider reads the config file automatically.

What Aider Can Do

Once running, Aider operates in a REPL. Drop a question or instruction:

> Write a Python function to find all primes up to N using the Sieve of Eratosthenes

> Add type hints and a docstring to the function in sieve.py

> Refactor this: [paste code or just name the file]

> Debug this error: IndexError: list index out of range

> Write pytest tests for the sieve function

Aider will:

Read relevant files from your repo automatically
Generate code changes
Show you a diff
Ask to apply and auto-commit with a sensible message

Exit: Type /exit or press Ctrl+C.

Customizing Aider's Behavior via Prompt Engineering

Aider uses a built-in system prompt optimized for code editing. You can augment it with your own instructions in .aider.conf.yml:

# .aider.conf.yml
openai-api-base: http://localhost:8000/v1
openai-api-key: fake
model: openai/Qwen3-0.6B
system-prompt: |
  You are a Python expert. Always include:
  - Type hints on all functions
  - Docstrings in Google style
  - Time and space complexity as inline comments
  Prefer readability over cleverness.

Or via a separate file:

# Write your system prompt to a file
cat > ~/coding-assistant-prompt.txt << 'EOF'
You are an expert Python engineer at a systems programming company.
- Always add type hints
- Include complexity analysis
- Prefer stdlib over third-party where possible
- Flag security issues when you see them
EOF

# Use it with Aider
aider --system-prompt ~/coding-assistant-prompt.txt

Why Prompt Engineering Matters

The system prompt shapes everything about how the model responds:

Technique	Effect	Example
Role Assignment	Sets behavioral context	"You are an expert in memory management"
Task Enumeration	Defines scope	"specializing in: Rust, C++, systems code"
Quality Markers	Influences style	"Always add tests"
Constraint Setting	Focuses behavior	"Never use eval(); flag unsafe patterns"
Response Format	Shapes output structure	"Include complexity analysis in comments"

These same techniques work for all LLMs — what you learn here applies to GPT, Claude, Gemini, or any hosted model.

One-Shot Mode (No REPL)

For quick questions without entering the REPL:

# Ask a question about a file
aider --message "Explain what this does and flag any bugs" mymodule.py

# Generate something from scratch
aider --message "Create a Flask app with a /health endpoint" --no-git

# Pipe code for a quick review
cat suspicious_code.py | aider --message "Review this for security issues" -

Comparing Approaches

Approach	When to Use
Aider + vLLM (this lesson)	Day-to-day coding; editing actual project files
Direct API (Lessons 4–5)	Learning tt-metal internals; building custom tools
vLLM chat interface	Quick Q&A; no file editing needed

Key insight: Aider's system prompt + your local Tenstorrent LLM gives you ~80–90% of a hosted coding assistant at zero API cost. The model runs on your hardware, your code stays private, and you can swap models in seconds.

Troubleshooting

"Model not found" or API errors:

Verify vLLM is running: curl http://localhost:8000/v1/models
Check the model name in the response matches what you passed to --model
The model name must match exactly: openai/Qwen3-0.6B where Qwen3-0.6B matches the model slug in vLLM's response

"No response" or very slow responses:

Check vLLM server logs for errors
Qwen3-0.6B is fastest; larger models take longer per token
Reduce context by starting Aider in a subdirectory rather than the repo root

Aider edits files unexpectedly:

Aider only edits files you've mentioned or it has inferred are relevant
Use --dry-run to preview changes without applying them
All edits are git commits — git log and git diff HEAD~1 let you review everything

N150 / QB2 model recommendation:

Use Qwen3-0.6B — it's fast, reasoning-capable, and fits comfortably in N150 DRAM
The prompting techniques here work identically regardless of model size

Key Takeaways

✅ Aider is zero-conf — one install, two env vars, then aider

✅ Your hardware, your data — the LLM runs on-device, nothing leaves your machine

✅ Prompt engineering is universal — the system prompt techniques here work with every LLM you'll encounter

✅ Persistent config — .aider.conf.yml in your repo makes it permanent

✅ All hardware supported — Qwen3-0.6B via vLLM works on N150, N300, P300c, and above