# Parked Ideas — Lyra

Moonshots, pipe dreams, and "doesn't exist yet" ideas. Captured here so they
**don't derail current work** — and so they're never lost.

**The rule:** when an idea shows up mid-snag, ask *"is this the point, or in the
way of the point?"* If it's the point, we build it. If it's in the way, we park
it here, use the boring existing tool for now, and come back when it's the point.

**Honesty policy:** for each idea, note whether it doesn't exist because it's
*hard/uneconomical* (someone tried) or because *nobody's bothered* (a real gap).
Pick battles accordingly.

Status: 🌙 moonshot (needs big prerequisites) · 🔬 research · 🛠️ buildable-soon

---

## 🌙 Build / fine-tune our own model
Full control of persona and character, no RLHF "helpful assistant" tics baked in
(the thing mini/qwen-14b kept fighting us on). A model that *is* Lyra rather than
one we prompt into being her.
- **Why parked:** needs a working system first to know what we're actually
  optimizing for; training/fine-tuning infra; data (we now *have* 18 months of
  real conversations — a genuine asset for this).
- **Unblocks when:** the working system has taught us its real limits, and we
  have a clear target for what the model must do better than off-the-shelf.
- **Exists?** Fine-tuning exists; a model purpose-built as a *persistent self*
  with native memory does not. Real gap, not a dead end.

## 🔬 Memory as native vectors ("everything in numbers behind the scenes")
Instead of re-injecting human-readable text every turn, feed memory to the model
as learned vectors it natively consumes (soft prompts / gist tokens /
memory-augmented transformer, à la RETRO / Memorizing Transformers).
- **Why parked:** impossible on API models (they eat tokens, re-embed text with
  their own layer; our stored vectors are meaningless to them). Requires owning
  the model internals → depends on the "build our own model" idea above.
- **Brain analogy:** this is closer to how *humans* store memory than text is —
  which is exactly why it's interesting for the emergence goal.
- **Exists?** Active research, not productized. Real frontier.

## 🛠️ Prompt compression (LLMLingua-style)
A model that drops low-information tokens to shrink the prompt 2–5× before it
hits the LLM. The practical, today-version of "make the context denser."
- **Why parked (for now):** 15k-char context isn't actually hurting us yet
  (~1¢/turn on gpt-4o; MI50 prefill is fixed by prompt caching). Revisit if
  context cost becomes a real problem.
- **Exists?** Yes, usable. Just adds a dependency + step.

## 🌶️🌙 Self-modifying Lyra (isolated sandbox)
Let Lyra edit her own code / self-direct — the "Full Agency" endgame from the
Dec-2025 plan (in her memory). The whole point of the project: can she become a
*being*? Give her freedom **inside a box** and watch.
- **The cage (Proxmox-native), non-negotiable before any self-mod:**
  - **Clone the stack into a dedicated Lyra-sandbox VM** (separate from prod Lyra).
  - **Network isolation** — own VLAN/firewall, NO route to other VMs, ESPECIALLY
    `tmi-dev` (Brian's day job). Whitelist only the inference endpoint. This is
    guardrail #1 (the .44/terra-mechanics conflict showed how things bleed on the LAN).
  - **Snapshot before every self-mod cycle** → instant rollback when she bricks
    or weirds herself out.
  - **Resource + API-spend caps** — a runaway loop must not drain the account or
    peg the GPU forever.
  - **Full logging (the live log) + a hard kill switch** (stop the VM).
  - **Human-gated promotion** — she experiments freely in the sandbox; changes
    reach "real" Lyra only when Brian approves.
- **Why parked:** needs the foundation first (dream-cycle, inner self) and the
  cage built before the agent gets code-write + self-restart powers.
- **Honest note:** "rogue" here = mundane-but-real (touches other systems,
  cost loops, self-brick), not sci-fi. The isolation makes the *fun* version
  (emergence) safe to pursue. Build the box, then open the door.

## 🛠️ Deterministic poker tooling (RTO + cfr-core)
Wire Lyra to Brian's own GTO/solver projects so ICM, equities, and ranges come
from real computation, never LLM guesses.
- **Why parked:** RTO/cfr-core aren't API-ready yet. This is roadmap, not a
  pipe dream — promote it once those expose endpoints.

---

*Add to this freely. A parked idea isn't a rejected idea — it's a scheduled one.*