1-800-HOTNITE / LA FIELD PROTOTYPE
scheduled outbound event concierge

THE CITY CALLS FIRST.

An LA night-out engine that learns what you actually go to, ignores what you hate, and moves you from “maybe” to a concrete plan.

Product model

This is not a generic calendar. It is a filtering, taste, and execution layer between a person and the overwhelming sprawl of LA event information.

1. Ingest

Scrape and import event candidates from local calendars, venue pages, newsletters, ticketing platforms, editorial sites, and manual operator finds.

2. Normalize

Convert messy listings into a standard object: title, venue, date, price, vibe, genre, door time, location, source, risk flags, ticket URL.

3. Classify

Tag events by scene, crowd, energy, social load, taste cluster, accessibility, cost, neighborhood, and likely user fit.

4. Rank

Score each option against the user’s actual behavior: likes, skips, hard no’s, distance tolerance, boredom threshold, and prior attendance.

5. Call

Outbound voice agent presents the shortlist one item at a time and listens for command words or normal human objections.

6. Execute

Text links now. Add deep links later. Introduce human-in-loop bookings before any autonomous purchasing.

Voice command behavior

The important design choice: the user is never trapped listening to an agent describe something they already know they hate.

HOTNITE: I’ve got five things that fit your radius and no-bro-EDM rule. First one is a small gallery opening in Chinatown, 7 to 10, free, likely low social pressure.
USER: NEXT.
HOTNITE: Skipping without penalty. Next: restored 35mm screening in Los Feliz, 9:30, not sold out, good fit for your repertory-film cluster.

LA source map

The advantage is coverage plus discrimination: official platforms for inventory, local weird calendars for texture, and manual additions for the stuff algorithms miss.

ScenestarIndependent LA music listings, concert calendar, contests, secret shows, set times.
Restless NitesHand-selected parties, concerts, comedy, and nightlife events.
Venue calendarsZebulon, Lodge Room, 2220, Dynasty Typewriter, Brain Dead, Aero, New Bev, etc.
Ticket platformsEventbrite, Dice, Ticketmaster, Fever, AXS, See Tickets, Resident Advisor where available.
EditorialEater LA, LAist, LA Times, DoLA, Infatuation, KCRW, art guides, museum calendars.
NewslettersSubstacks, promoter blasts, gallery emails, venue weekly digests.
Social/manualInstagram flyers, promoter posts, artist pages, friend-of-friend discoveries.
Places layerMaps/hours/address validation, not taste authority.

Taste intake

The user profile should be mostly tappable and voice-editable. Start broad, then learn through “NO,” “NEXT,” saves, and attendance.

WANTS
repertory filmsmall showsgallery openingsnew restaurantsdancingcomedypickup gameslecturesbars with character
HARD NO
bro EDMbottle servicenetworkingwellness griftcrowd work comedyinfluencer restaurantsstanding in lineparking nightmare
SOCIAL LOAD
solo-friendlydate-friendlybring friendslow-pressuretalk to strangersanonymous in crowd
ENERGY
quietstrangelateromanticphysicalcerebralcheapnearby

Recommendation demo

Mock data only. The point is the shape of the object and how fit is explained without sounding like a marketing blurb.

Current budget: $35. Use the buttons on the card to simulate voice feedback.

Build sequence

Keep it small enough to produce. The first version should feel agentic while remaining operationally safe.

01Static site

Signup form, taste intake, phone/SMS consent, LA-only waitlist.

02Event database

Supabase tables for sources, events, venues, tags, users, feedback.

03Operator console

Approve/reject scraped events, add taste tags, mark reliable sources.

04Voice test

Twilio outbound call with command parser and SMS recap.

05Execution layer

Links first. Deep links second. Human concierge third. Autonomous booking last.

Data objects

The actual backend should revolve around these records.

ObjectFieldsWhy it matters
Eventtitle, date, venue_id, price, source_id, URL, age limit, sold-out status, raw description, normalized summaryStandardizes listings from totally different sources.
Venuename, address, neighborhood, coordinates, transit notes, parking pain, scene tagsPrevents the agent from recommending impossible logistics.
Taste taggenre, crowd, social load, energy, scene, format, novelty, bro-risk, normie-riskSeparates “electronic music” from “wrong electronic music.”
User profilehome zone, radius, budget, preferred call time, wants, hard no’s, mobility, attendance historyMakes the outbound call feel personally useful instead of spammy.
Feedbackevent_id, user_id, command, reason, timestamp, call_idTurns NO/NEXT/GO BACK/REPEAT into taste memory.
Call sessionqueue, current index, prior index, transcript, selected event, SMS sentSupports GO BACK and REPEAT without confusion.