Practical notes: Building Local AI Agents: A Practical Guide to Models, Memory

This walkthrough rebuilds the path from raw materials to a working system for: Building Local AI Agents: A Practical Guide to Models, Memory, and Orchestration. The focus is operable steps, explicit checks, and code that you can drop into a repo without guessing intent. For the Overview stage, define the inputs, the owner of the step, and the exit criteria before changing code. Operators should be able to re-run the step from a known checkpoint without guessing hidden state. Treat this stage as a contract between inputs and validated outputs. Name the artifacts, define success checks, and refuse silent partial completion.

Layer 1: LLM Layer

When working through the Layer 1 LLM Layer stage, write down the contract first: required inputs, success signal, and what happens on partial failure. That checklist keeps later code changes honest. Record timings and token or query cost next to functional results. Cost visibility early prevents surprise bills when the path moves from demo to shared environments. Cache stable system instructions and tool schemas. Re-sending identical preamble is a common source of burn.

ollama pull qwen3:8b
ollama pull nomic-embed-text

_orig_ollama_chat = ollama.chat
def _no_think_chat(*args, **kwargs):
    opts = kwargs.get("options") or {}
    if isinstance(opts, dict):
        opts.setdefault("think", False)
        kwargs["options"] = opts
    return _orig_ollama_chat(*args, **kwargs)
ollama.chat = _no_think_chat

Layer 2: Agent Framework Layer

When working through the Layer 2 Agent Framework stage, write down the contract first: required inputs, success signal, and what happens on partial failure. That checklist keeps later code changes honest. Keep configuration outside application code. Environment files, secret stores, and feature flags belong in one place operators can audit without reading the whole graph. Cache stable system instructions and tool schemas. Re-sending identical preamble is a common source of burn.

- -
name: local-ai-assistant
description: Local AI coding assistant with persistent memory
command-dispatch: tool
command-tool: exec
command-arg-mode: raw
-

def detect_intent(message: str) -> dict:
    try:
        resp = ollama.chat(
            model=OLLAMA_CHAT_MODEL,
            messages=[{"role": "user", "content": INTENT_PROMPT.format(message=message)}],
            options={"temperature": 0, "num_predict": 1024},
        )
        return extract_json(resp["message"]["content"])
    except Exception:
        return keyword_intent_fallback(message)

Layer 3: Memory Layer

When working through the Layer 3 Memory Layer stage, write down the contract first: required inputs, success signal, and what happens on partial failure. That checklist keeps later code changes honest. Document the happy path and the recovery path together. Retries, human gates, and dead-letter handling are part of the product, not later polish. Cache stable system instructions and tool schemas. Re-sending identical preamble is a common source of burn. When working through the Layer 3 Memory Layer stage, write down the contract first: required inputs, success signal, and what happens on partial failure. That checklist keeps later code changes honest. Treat this stage as a contract between inputs and validated outputs. Name the artifacts, define success checks, and refuse silent partial completion.

from mem0 import Memory
config = {
    "llm": {
        "provider": "ollama",
        "config": {"model": "qwen3:8b", "ollama_base_url": "http://localhost:11434"},
    },
    "embedder": {
        "provider": "ollama",
        "config": {"model": "nomic-embed-text", "ollama_base_url": "http://localhost:11434"},
    },
    "vector_store": {
        "provider": "qdrant",
        "config": {"host": "localhost", "port": 6333, "embedding_model_dims": 768},
    },
}
memory = Memory.from_config(config)
memory.add("I always use type hints and pytest", user_id="dev")
results = memory.search("write a utility function", user_id="dev")

def _is_worth_storing(self, user_message: str) -> bool:
    response = ollama.chat(
        model=OLLAMA_CHAT_MODEL,
        messages=[{"role": "user", "content": SMART_MEMORY_PROMPT.format(
            user_message=user_message
        )}],
        options={"temperature": 0, "num_predict": 512},
    )
    data = self._extract_json_robust(response["message"]["content"])
    return bool(data.get("worth_storing", False))

HIGH-VALUE (worth storing):
- "I prefer TypeScript over JavaScript"
- "I use pytest, never unittest"
- "Always use Google-style docstrings"

LOW-VALUE (discard):
- "What does enumerate() do?"
- "Write a retry decorator"
- "Thanks"

Layer 4: Storage Layer

The Layer 4 Storage Layer stage works best when treated as a measurable surface. Capture one golden transcript, one failure case, and the rollback note before expanding scope. Record timings and token or query cost next to functional results. Cost visibility early prevents surprise bills when the path moves from demo to shared environments. Budget tokens per turn and per session. Agentic tools expand context aggressively; hard caps keep demos from becoming surprise invoices.

pip install chromadb

# Qdrant via Docker
docker run -d - name qdrant-local -p 6333:6333 \
-v $(pwd)/qdrant_storage:/qdrant/storage qdrant/qdrant

"vector_store": {
    "provider": "qdrant",
    "config": {
        "collection_name": "coding_assistant",
        "host": "localhost",
        "port": 6333,
        "embedding_model_dims": 768,  # must match your embedding model exactly
    },
}

Layer 5: Interface Layer

The Layer 5 Interface Layer stage works best when treated as a measurable surface. Capture one golden transcript, one failure case, and the rollback note before expanding scope. Keep configuration outside application code. Environment files, secret stores, and feature flags belong in one place operators can audit without reading the whole graph. Budget tokens per turn and per session. Agentic tools expand context aggressively; hard caps keep demos from becoming surprise invoices.

Wrapping Up

The Wrapping Up stage works best when treated as a measurable surface. Capture one golden transcript, one failure case, and the rollback note before expanding scope. Document the happy path and the recovery path together. Retries, human gates, and dead-letter handling are part of the product, not later polish. Budget tokens per turn and per session. Agentic tools expand context aggressively; hard caps keep demos from becoming surprise invoices. The Wrapping Up stage works best when treated as a measurable surface. Capture one golden transcript, one failure case, and the rollback note before expanding scope. Treat this stage as a contract between inputs and validated outputs. Name the artifacts, define success checks, and refuse silent partial completion.

Interface: Web UI on localhost
│
Framework: OpenClaw - tool-dispatch, SKILL.md routing
│
LLM: Ollama for qwen3:8b
(think mode disabled for JSON reliability)
│
Memory: Structured persistent memory with value filtering
Embeddings via nomic-embed-text
│
Storage: Chroma (zero infrastructure) or Qdrant (more robust)

Operational checklist

The Operational checklist stage works best when treated as a measurable surface. Capture one golden transcript, one failure case, and the rollback note before expanding scope.

Prefer small, testable units over sprawling scripts. When a step fails, the failure should point at a single responsibility rather than a tangled pipeline.

Budget tokens per turn and per session. Agentic tools expand context aggressively; hard caps keep demos from becoming surprise invoices.

Put human approval on edges that spend money or change production data. Compile-time wiring does not equal business completeness.

Write a short runbook: how to rotate keys, how to drain the queue, how to roll back the last ingest.

Keep configuration outside application code. Environment files, secret stores, and feature flags belong in one place operators can audit without reading the whole graph.

Before promoting the stack, freeze versions, capture a golden transcript for the critical path, and confirm rollback steps. Shared environments need rate limits, tenancy checks, and a clear owner for secret rotation. Prefer boring reliability over clever one-off demos.

Batch note for 12622e9e0269: keep provider keys out of the repo, set a per-session token ceiling, and store transcripts next to the eval fixtures so later model swaps stay comparable.

When working through the hardening note 0 stage, write down the contract first: required inputs, success signal, and what happens on partial failure. That checklist keeps later code changes honest. Prefer small, testable units over sprawling scripts. When a step fails, the failure should point at a single responsibility rather than a tangled pipeline.

Hardening detail 0/721: measure wall time, error class, and token spend for this note, then decide whether to keep the change based on a fixed question set rather than anecdote.

The hardening note 1 stage works best when treated as a measurable surface. Capture one golden transcript, one failure case, and the rollback note before expanding scope. Record timings and token or query cost next to functional results. Cost visibility early prevents surprise bills when the path moves from demo to shared environments.

Hardening detail 1/721: measure wall time, error class, and token spend for this note, then decide whether to keep the change based on a fixed question set rather than anecdote.

For the hardening note 2 stage, define the inputs, the owner of the step, and the exit criteria before changing code. Operators should be able to re-run the step from a known checkpoint without guessing hidden state. Document the happy path and the recovery path together. Retries, human gates, and dead-letter handling are part of the product, not later polish.

Hardening detail 2/721: measure wall time, error class, and token spend for this note, then decide whether to keep the change based on a fixed question set rather than anecdote.

When working through the hardening note 3 stage, write down the contract first: required inputs, success signal, and what happens on partial failure. That checklist keeps later code changes honest. Treat this stage as a contract between inputs and validated outputs. Name the artifacts, define success checks, and refuse silent partial completion.

Hardening detail 3/721: measure wall time, error class, and token spend for this note, then decide whether to keep the change based on a fixed question set rather than anecdote.

The hardening note 4 stage works best when treated as a measurable surface. Capture one golden transcript, one failure case, and the rollback note before expanding scope. Keep configuration outside application code. Environment files, secret stores, and feature flags belong in one place operators can audit without reading the whole graph.

Hardening detail 4/721: measure wall time, error class, and token spend for this note, then decide whether to keep the change based on a fixed question set rather than anecdote.

For the hardening note 5 stage, define the inputs, the owner of the step, and the exit criteria before changing code. Operators should be able to re-run the step from a known checkpoint without guessing hidden state. Prefer small, testable units over sprawling scripts. When a step fails, the failure should point at a single responsibility rather than a tangled pipeline.

Hardening detail 5/721: measure wall time, error class, and token spend for this note, then decide whether to keep the change based on a fixed question set rather than anecdote.

When working through the hardening note 6 stage, write down the contract first: required inputs, success signal, and what happens on partial failure. That checklist keeps later code changes honest. Record timings and token or query cost next to functional results. Cost visibility early prevents surprise bills when the path moves from demo to shared environments.

Hardening detail 6/721: measure wall time, error class, and token spend for this note, then decide whether to keep the change based on a fixed question set rather than anecdote.

The hardening note 7 stage works best when treated as a measurable surface. Capture one golden transcript, one failure case, and the rollback note before expanding scope. Document the happy path and the recovery path together. Retries, human gates, and dead-letter handling are part of the product, not later polish.

Hardening detail 7/721: measure wall time, error class, and token spend for this note, then decide whether to keep the change based on a fixed question set rather than anecdote.

For the hardening note 8 stage, define the inputs, the owner of the step, and the exit criteria before changing code. Operators should be able to re-run the step from a known checkpoint without guessing hidden state. Treat this stage as a contract between inputs and validated outputs. Name the artifacts, define success checks, and refuse silent partial completion.

Hardening detail 8/721: measure wall time, error class, and token spend for this note, then decide whether to keep the change based on a fixed question set rather than anecdote.