1/10

The studio desk at night; the centre monitor shows Bishop’s panel.
The room from the homepage, with the centre monitor, Bishop’s screen, lit.

02 · Bishop · planned: a hands-free Googler

Became a living bot, with humor and life.

A voice AI that sits in my friends’ Discord calls: he listens, talks back, and remembers.

Bishopin a callSeattle

He hears, thinks, then responds.

in a call
content

Bishoplistening

  1. Hearsspeech becomes text, per friend
  2. Thinksthe AI decides: is this a moment for me?stay quiettalk
  3. Speakstext becomes his voice

“Tess wins. Rowan, sit down.”

He has a whole day of his own.

His daygenerated17 Sep · wake 07:35 · sleep 01:45
  1. 09:50walk to library, refill water
  2. 10:15half-reading notes, phone winning
  3. 11:45chicken wrap and cold brew
  4. 12:20drive home
  5. 12:45quick nap
  6. 13:10laundry, tidy kitchen
  7. 13:40gas station run
energycoffee is a loan
7 of the day’s 23 blocks, from his database · also on the clock: class Mon/Wed, a show, a game, and pop-ups like “a fade at the barber” (daysim.py) · energy line illustrative

a new day planned for every date, on the real clock · 17 Sep 2026

He remembers you. Weeks later.

Memoryepisodes

“the exam thing”

a few weeks ago · afternoon

The MCAT she wouldn’t stop worrying about

Tess “if I bomb this I’m dropping out of pre-med lol”
written up that night, the real lines kept
Days later, unpromptedbrought it up himself
  1. Bishop → Tess · 6 days laterhey wait, did you ever hear back on the MCAT?
nobody asked him to remember. he just did.

His memory comes in nine parts.

Memory9 stores

Systems

  • Self-canon: who he is, to himself
  • Hard facts: a locked biography, owner-approved
  • Nightly digest: the day, written up as stories
  • Live transcript: the last 15 minutes, word for word

Reading people, 5 layers

  • Favor: a number for how he feels about you
  • Impressions: things he noticed, each with a quote
  • Traits: claims that firm up with evidence
  • Dyads: how two OTHER people get along
  • Profiles: his summary of you, in his own words
4 systems + 5 layers of people · 17 Sep 2026

While you sleep, he writes the night down.

4:00 am · the callraw transcript
  • Tess if I bomb this I’m dropping out of pre-med lol
  • Rowan you’ve said that before every exam
  • Bishop when is it?
4:01 am · written upsaved

The MCAT she wouldn’t stop worrying about

Tess “if I bomb this I’m dropping out of pre-med lol”

at 4am, while everyone is asleep · 726 conversations written up across 105 days · 17 Sep 2026 · the lines are invented

He reads how you two click.

The crew, to himrelationship web
BBishop
RRowan
JJuniper
TTess
MMarcus
DDevin
BBishop
RRowan
JJuniper
TTess
MMarcus
DDevin
a view he formedfriction he sawoften together, no view yeta group he's modelled
Rowan & Junipera pair, not a person
  • he’d expect: trash talk about last night’s gamepredictable
  • would surprise him: Juniper asking for real adviceunlikely
his opinion, invented for illustration. Colors are tone only, never whether he helps

Mood, energy, and how much he shares.

Moodfrustrateddecided, not rolled
in a callcoffee, on loansleep pays it back8a12p4p8p12a

Something real happened, so this one holds for hours instead of the usual 55 minutes.

with Rowan · close

“ngl that lag spike pissed me off”

with someone he just met

keeps it to himself

illustrative. How close you are decides what he shares, never whether he helps you

One person built all of it.

56,092lines of Python code

105days of his own simulated life

73.4%of the times he could act on his own, he picks: do nothing

counted September 2026

recreation

Bishop’s real admin panel shows private conversations, so it sits behind a login. This is its interface rebuilt for this page: the same sky, layout and type, with stand-in names (Rowan, Juniper, Tess, Marcus, Devin). The call, the memory moment, his opinions of the crew and the mood moment are invented; his day, the numbers and the code are real.

Every number on this page, with the date it was counted

825 → 792 → 33 moments past every clock throttle → blocked by reading the room → he spoke
17 Sep 2026
13 named stages in the engage gate
17 Sep 2026
2–5 a day times he stops to think, in the gaps of his day
17 Sep 2026
73.4% of those thoughts end in “nothing”, out of 18 decisions he can make
17 Sep 2026
105 days of generated daily life, one plan per date
17 Sep 2026
9 memory stores (4 systems + 5 layers of people)
17 Sep 2026
14 · 12 command verbs · the ones that have actually fired
17 Sep 2026
15 behavioural metrics on control charts
26 Aug 2026
62 · 56,092 Python modules · lines (wc -l)
22 Sep 2026
2,235 / 2,235 self-test checks passing at boot, 67 suites, ~10.3 s, local run
22 Sep 2026
11 / 17 background loops / slash commands
22 Sep 2026
~90 named prompt layers, ~50,000 characters
22 Sep 2026
14 tabs in the real admin panel
22 Sep 2026
726 · 13 · 105 conversations · people · days, one box
17 Sep 2026

The self-test runs as a plain Python process at boot. Line counts are raw wc -l.

What measuring found: seven problems, none of which crashed

  1. 2,606 ms

    A 1.7 µs lookup froze him for 2,606 ms: one shared SQLite connection stalled the event loop. 4,000 reads under write load: 12.813 s blocked, 0.004 s with a connection per thread (~3,200× less). Its link to a routing gap is a hypothesis with good evidence, not a proof. BISHOP_AUDIT_2.md:4749-4776 · 17 Sep 2026

  2. 82%

    of what he recalled was about his own name: 289 of 351 lines. The search matched his name, not the topic. Fixed on the query side. BISHOP_AUDIT.md:1440-1441 · 26 Aug 2026

  3. 12×

    A text-similarity check got an exact upper bound: 10,461 of 10,464 pairs rejected before any real work, 945 → 79 ms p50. Measured in isolation; the effect in production is unmeasured. BISHOP_AUDIT_2.md:4742-4745 · 16 Sep 2026

  4. 3 → 8

    A yes/no gate’s token budget. The next model returned nothing at 3, three draws of three, and nothing would have errored. router.py · 22 Sep 2026

  5. def, def

    The weekly audit defined one check twice; Python silently kept the second. A self-test now forbids it. selftest.py

  6. 0 / 15

    “0 entries” was fifteen failed queries: 67 rows still named a removed person. Every scan now reports its success count. BISHOP_AUDIT_2.md:4791-4795 · 17 Sep 2026

  7. 17%

    of live turns misfiled by an unregistered prompt layer, live from 31 Jul, found 21 Sep 2026. trace_bishop.py:1343-1358

One sentence, from separate microphones to a reply

  1. Voice packets
  2. A buffer per speaker
  3. A recognizer per speaker
  4. Turn detection
  5. Coalescer
  6. Engage gate
  7. Planner
  8. Feeders (6 s budget)
  9. Prompt assembly
  10. Reply + guards
  11. Speech

Most sentences stop at the engage gate: 13 named stages, cheapest first. BISHOP_AUDIT_2.md:126-153

The budgetbishop26/router.py:17-29
# A ONE-WORD GATE STILL NEEDS HEADROOM, AND 3 IS NOT IT ON EVERY MODEL.
# These are YES/NO gates, so `max_tokens=3` looked obviously sufficient and was, on
# gemini-2.5-flash-lite. Measured Sep 11 against its named successor, 3 draws per budget:
#
#   gemini-3.1-flash-lite   mt=3 ['<EMPTY>','<EMPTY>','<EMPTY>']   mt=5 ['YES','YES','YES']
#   gemini-2.5-flash-lite   mt=3 ['NO','NO','NO']                  mt=5 ['NO','NO','NO']
#
# It is NOT reasoning tokens eating the budget (`reasoning_tokens` is 0 on both) -- 3.1 simply
# returns nothing that small. Every one of these gates falls back to a default on an empty
# reply, so the whole failure would have been three gates silently answering "default" forever,
# with no error anywhere. `gemini-2.5-flash-lite` retires Oct 16 2026, so this WILL be swapped.
# 8 is the same price in practice (a gate emits 1 token) and survives the next model too.
_GATE_TOKENS = int(os.getenv("BISHOP_GATE_TOKENS", "8"))
The measurement stays beside the fix, so nobody lowers the budget back to 3.
The spread: 15 metrics on control chartsbishop26/review.py:644-662
def _sigma(vals):
    """Spread from the mean MOVING RANGE, not the standard deviation.

    This is the individuals-chart estimator, and here it earns its keep twice over. The stated risk
    for this whole feature is a baseline drawn from an already-drifted stretch, which bakes the
    drift in -- and a standard deviation absorbs any trend inside the baseline into the spread, so
    the limits get widest exactly when behaviour was already sliding. Point-to-point variation
    doesn't see a slow slide, so it stays honest about how much this number normally jumps."""
    vals = [v for v in vals if v is not None]
    if len(vals) < 2:
        return None
    mr = [abs(b - a) for a, b in zip(vals, vals[1:])]
    s = (sum(mr) / len(mr)) / 1.128          # d2 for n=2
    if s > 0:
        return s
    try:
        return _stat.pstdev(vals) or None    # identical neighbours but a spread overall
    except Exception:
        return None
The band comes from the moving range, so a baseline that was already drifting cannot widen its own limits.
Everything he can dobishop26/llm.py:1305-1306
EXECUTABLE_COMMANDS = ("KICK", "TEXT", "SAYCALL", "JOINCALL", "LEAVE", "DRAG",
                       "TIMER", "REMIND", "CHOOSE", "MUSIC", "GAME", "IMAGE", "PLAN", "QUEUE")
One tuple. The parser and the trace recorder both import it, so they cannot drift apart. Source private; excerpts verbatim.

What I don’t claim

  • Scale: 726 conversations, 13 people, 105 days, one box. Friends, not users.
  • Four model roles in code; two models in the live container.
  • No latency figure, because none is current.
  • Psychology papers are design sources, not validation.

Measured, and not built

  • A sarcasm detector: its best signal also describes ordinary speech.
  • A memory fix: it passed offline, then got worse against the real index. Reverted.

BISHOP_AUDIT_2.md:4517-4572 · 17 Sep 2026

Stack

One Python process, in Docker, on one Linux box.

py-cordDeepgram FluxOpenRoutermem0 + embedded QdrantElevenLabsTavilySQLite (WAL)Dockerasyncio

The panel itself

FastAPI + React 19 + Tailwind 4 + Vite, 14 tabs, read-only mounts of his data. Private, behind Cloudflare Access.

Proves real-time multi-speaker voice pipeline, nine-store memory, observability for a box with no interactive debugging (résumé → Bishop).