schultz.ai Labsdeveloper platform
A platform built on my brain.
A developer platform sitting on a structured model of how I make consumer-AI product calls — real primitives, grounded answers, self-scored. Here’s the live endpoint. Hit it.
A live command line on my brain — every command runs against the real record, grounded and cited.
02The apps
16 apps · all live
Apps built on the platform.
Every one runs on the same brain — the corpus to ground it, the judge to score it. Open any of them.
The Judgment Call
NewBring a real, thorny consumer-AI decision. Get a grounded, opinionated take cited to Robert's actual record — honest where his record doesn't cover you, and self-scored for grounding before it speaks.
Hiring Panel
Convene a panel of AI interviewers on Robert's record — each probes from its own lens and returns a structured hire / no-hire verdict you can share.
Evals
The persona AI grades its own answers — groundedness, accuracy, helpfulness, calibration — live, in production. A site about honest quality that measures its own.
Insights
Real telemetry of the persona AI — traffic, conversion funnels, tool calls, and quality trends from PostHog + Neon.
Studio
The developer studio. Write code against the brain SDK and run it live — hello-world apps, grounded questions, opinion calls, fit checks, and evals, in a sandboxed editor with the full API reference.
Cold Open
Drop a role or a company. The brain weighs the fit against Robert's actual record — honest about the stretch — and drafts the short, cited note he'd actually send.
Two Truths & a Lie
A game on the anti-hallucination spine: the brain deals two real, cited career facts and one plausible fake it made up on purpose. Spot the lie — then see the receipts.
Career Constellation
The whole brain as a picture: pick a skill and watch it draw a constellation across the roles it actually spans — with the real receipts underneath. Instant, client-side, no model call.
The Briefing Room
Pick a lens and the brain scripts a two-voice deep dive on Robert's record — a curious host and a skeptic who pushes on the gaps — then reads it back in two distinct voices, highlighting each cited line as it speaks.
Remix
Ask one question and the brain answers it four honest ways — the Operator, the Eval Purist, the Platform Builder, the 0→1 Device Lead — each a cited take with its own grounding score. Remix for new angles, upscale any into a full argument.
The Anthem
Name a topic and pick a vibe; the brain writes a song about Robert's career — every lyric true to the record — then records it into an actual track you can play. Grounded lyrics, real audio.
Time Machine
Drag or play through the years and watch Robert's career unfold — the voice → multimodal → ambient progression lighting up, with the real role and the real numbers at every point. Instant, client-side, no model.
Live Roles
Pulls this month's 'Who is hiring?' postings live from Hacker News (no key), then the brain ranks the senior AI / product / leadership openings by honest fit to Robert's record — each with a fit score, the cited why, and the single biggest gap. A real 3rd-party integration on the fit() spine.
Frontier Watch
Pulls the leading arXiv research on a frontier-AI topic (live, no key) and the brain maps each paper onto Robert's real record — shipped it, worked adjacent, or it's ahead of where he's been — with an honest one-line read and the receipts. Frontier → product, made legible.
Deep Dive
Ask one hard question and watch a real agent work it: the brain plans the lines of inquiry, researches each against the record live, then synthesizes one grounded answer — and self-scores its own groundedness before it lands. The anti-hallucination spine, made visible.
Screen Test
Name a topic and Robert delivers a short take to camera: the brain writes a script true to the record, his cloned voice reads it, and a video model animates a reference photo to deliver it — lip-synced to his actual voice. A grounded talking-head video, clearly labelled AI-generated.