Data + AI consulting
Senior pair-hours on your data and AI work — source selection, scoring logic, prompt and eval design, and a second set of eyes on what to trust. Billed by the hour.
A studio that watches the data sources that matter, scores what's important with AI, and turns the noise into plain-English alerts — the signal your team needs, the moment it appears.
Tiny Prism LLC is a one-person studio registered in Sheridan, Wyoming. Two decades building for the web, the recent years spent turning sprawling data into AI-scored, plain-English alerts. We build our own monitoring products — and the same systems for one or two teams at a time.
Founder · Data + AI engineer
Building for the web since 2005, fully remote with US and EU teams since 2011 — now focused on data monitoring, AI analysis, and alerting.
The work follows one pattern, the one our own products run on: watch a data source — a public register, an API, your own database — detect what changed, score what matters with an LLM, and deliver it as an alert someone will actually read. Less dashboard, more decision.
What you actually get: a working slice within a week — a first scored record or a sample alert — then steady, demoed progress instead of status decks. A runbook and an architecture note when we're done, so the next person to touch it can.
If something is outside our focus, we'll tell you within a day and point you toward someone better suited. We'd rather pass than half-ship.
Every engagement runs on the same loop — monitor, analyze, alert. Anything outside it, we'll tell you upfront and point you to someone better suited.
Senior pair-hours on your data and AI work — source selection, scoring logic, prompt and eval design, and a second set of eyes on what to trust. Billed by the hour.
Watch any source — public registers, APIs, your own database — detect what changed, and deliver scored alerts, by digest or in realtime. The engine behind our own products.
LLMs pointed at your data: plain-English summaries, relevance scoring, structured extraction, and evals — grounded in the source and always linked back to it.
Ingest, dedupe, entity-match, and normalize from feeds, APIs, and databases — the unglamorous plumbing that makes the data, and the alerts on top of it, trustworthy.
An end-to-end monitoring product like the ones we run in-house — watchlists, scoring, intelligence pages, alerts, and billing — built, shipped, and yours.
Send a brief paragraph. We'll tell you within a day whether it's a fit, and what it might cost.
Send a briefEverything we build sits on two capabilities — moving and shaping data, and pointing AI at it. The rest is in service of those two.
Continuous watch over public records and private feeds — ingested, deduped, entity-matched, and scored — then delivered as a digest or a realtime alert the moment something changes.
ingestion · entity-matching · digests · realtime alertsLLMs that read a filing, a notice, or a record and return a plain-English summary, a relevance score, and the reason it matters — grounded in the source, with the latest OpenAI, Claude, and Gemini models.
summaries · scoring · extraction · evalsAlso in the kit — public-record ingestion · fuzzy & phonetic matching · relevance scoring · source-linking · CSV / API export · daily & realtime delivery
Every project runs on the same cadence. You'll always know what's next, what we're blocked on, and what's shipping this Friday.
An hour on a call, then a day in your repo and your docs. You get back a one-page plan with a fixed quote and a start date.
A real, working slice — often a first scored record or a sample alert against your own data. A visible stake in the ground that sets the cadence for everything after.
Every Friday: a 15-minute recorded walkthrough of what shipped, what's in review, and what's next. No status decks, no theatre.
Runbook, test suite, and an architecture note you can hand to your next hire. Thirty days of follow-up included, whether you need it or not.
A recent engagement — typical scope, typical pace. Client names are withheld by default; references and fuller write-ups are available on request.
A consumer marketplace for scheduled, payable lessons — where every booking, payout, and refund had to reconcile. One engineer owned the data and reporting across releases and integrations: keeping numbers trustworthy, surfacing what changed, and making sure the money and the records always agreed.
Ask for the write-upLLM drafting and editing flows — prompts, retries, structured output, and evals.
ai · llm · summarization 2023 – 24 05Featured above. Multi-year ownership: payouts, refunds, reconciliation, reporting.
data · reporting · reconciliation 2022 – 24 04A content pipeline publishing into Google Docs, Medium, HubSpot, and WordPress.
pipelines · apis · automation 2021 – 22 03A multi-retailer marketplace — catalog data, storefront, and admin reporting.
marketplace · catalog · admin 2021 – 23 02Patients and clinicians reviewing device data — charts, configuration, exports.
device data · charts · exports 2019 – 20 01Dashboards for wallet usage statistics and operational reporting.
analytics · dashboards · sql 2018 – 19Our products are the clearest statement of what we do. Each one runs the same Data + AI loop we build for clients: watch an official public record, score what actually matters, and send a plain-English alert linked to its source — running live, in production.
Find the right federal opportunity before your competitors do.
Monitors SAM.gov and USAspending, summarizes every new opportunity in plain English, and sends only the alerts that match your NAICS, agencies, and competitors.
The SEC filings that actually move stocks — in plain English.
Watches every insider buy, activist stake, and 8-K on SEC EDGAR, scores the ones that matter, and sends a plain-English alert the moment they're filed.
Know the moment a similar trademark appears.
Monitors USPTO records and alerts you in plain English when a new application may conflict with your brand — risk scored, with the official record one click away.
Hourly when you have the team and just need senior eyes on your data or AI. Fixed-scope projects when you need a monitoring or alerting system built end-to-end. No retainers, no minimum contract length past the first ten hours.
Senior pair-hours on your data and AI work. Best when you have the team — you just need a second set of experienced eyes.
A scoped deliverable — a monitoring pipeline, an AI summarization layer, or a full data product. Fixed scope, fixed bid, weekly demos.
A paragraph is plenty — the data you want monitored and the alert you wish landed in your inbox. We read every brief and reply within one business day with a yes, a no, or a referral to someone better suited.