3h 48m 35s logged
Added a little class of travel indicator to help guide the budget further… working on displaying raw price data clearly and cleanly within the result now
8h 26m 42s logged
Took some time to learn how to link google flights, and have an accurate est. cost for trip options. Also am trying to implement some convinience measures to make input easier (multiple choice, checkboxes, etc.)
3h 8m 2s logged
I’ve got soemthing working here. Going to implement a little more practicality in terms of finding fares and using real world pricing.
1h 36m 3s logged
Got a nice UI going, trying to test.
4h 30m 37s logged
Working on the initial server, trying to make it API-less…
Ship
Pending review
AI-powered startup idea analyzer. Given a one-line pitch, it returns a structured analysis: competitors, market size, weaknesses, pricing suggestions, and a computed likelihood-of-success score.
Features
Likelihood of success, computed deterministically from sub-scores (not just an LLM guess)
Monetization score and technical difficulty rating
Market size and growth rate, verified against real published figures where possible
Competitor list with web-scraped details (website, description, pricing hints)
Weaknesses and risk severity
Pricing model suggestions with rationale
Tech Stack
Backend: Flask (Python)
AI: Groq API (Llama 3.3 70B)
Market data: scraped search results (Mojeek), parsed for dollar figures and CAGR
Competitor scraping: BeautifulSoup, Requests
Frontend: HTML, CSS, JavaScript
18m logged
Shipping now!
15m 25s logged
UI issue on the vercel, fix then push.
7h 16m 42s logged
I added sources so you can tell where it got the data from, the issue was that edge and chrome were rejecting requests so I had to go a seperate route
Found an issue with the model will all of the characteristics clustered around 50… Will need to fixthat.
5h 40m 32s logged
Implemented other insights and increased the different types of monetization models the model is trained on so it can make the best decision.
3h 50m 34s logged
Working on implementing the mathematics into my model. So far so good, I will now look at improving other insights and adding sources…
4h 30m logged
Did some math, to calculate the success index. I realized that the LLM returns the ingredients (market size, growth, monetization, difficulty, competitor count, weaknesses).
So I have to normalize the signals from 0-1,
market = clamp( (log10(TAM_dollars) - 7) / 4 , 0, 1 )
growth = clamp( CAGR% / 30 , 0, 1 )
n = number of competitors
n = 0 → 0.40 (unvalidated market)
n = 1–3 → 0.90 (validated, room to enter)
n = 4–6 → 0.70
n = 7–9 → 0.50
n ≥ 10 → 0.30 (saturated)
feasible = (10 - difficulty) / 9
success% = round( 100 / (1 + e^(-6 * (S - 0.5))) )
Now working on implementation
6h 29m 59s logged
Working on this validator, so far I have used knowledge from previous projects to get the website scraping to work, I am just not trying to get the predictions to work… Very rough right now, it may take a while for that to start working.
36m 26s logged
Starting to devlop this startup validator to give people a glimpse into how well their ideas will do on the market. Inspired by the bad sharktank pitches.
Ship
@jayeshsinghal
on
On Demand Judge
AI Debate Judge
A real-time multiplayer debate platform with AI-powered judging using Groq's Llama and Whisper models.
Features
Multiplayer Debates: Two players debate in real-time with WebRTC video streaming
Live Video: See your opponent during the debate with side-by-side video feeds
Speech-to-Text: Automatic transcription using Groq Whisper API
AI Judging: Comprehensive debate analysis using Groq Llama 3.3 70B
Public Forum Format: 8 speeches across 4 rounds (Constructive, Rebuttal, Summary, Final Focus)
Real-time Sync: Firebase Realtime Database for instant synchronization
Tech Stack
Pure HTML/CSS/JavaScript (no build tools)
Firebase Realtime Database
WebRTC for peer-to-peer video
Groq API (Whisper + Llama 3.3 70B)
Deployed on Vercel
How to Use
Visit ondemandjudge.vercel.app
Create a debate room or join with a room code
Wait for both players to connect
Start the debate and speak when it's your turn
Each speech has a 5-minute time limit
After all 8 speeches, get AI-powered judging results
Debate Format
Round 1: Constructive Speeches (Pro then Con)
Round 2: Rebuttal Speeches (Pro then Con)
Round 3: Summary Speeches (Pro then Con)
Round 4: Final Focus (Pro then Con)
Judging Criteria
The AI judge evaluates both teams on:
Claims Quality (0-10)
Evidence Strength (0-10)
Persuasiveness (0-10)
Winner is determined by overall score and detailed reasoning.
@jayeshsinghal
on
On Demand Judge
1h 54m 3s logged
Finalized the project by redoing the UI and fixing small streaming issues
@jayeshsinghal
on
On Demand Judge
3h 32m 44s logged
Redid the UI, fixed video streaming!
Tested it without speaking, and the judging seems pretty good!
@jayeshsinghal
on
On Demand Judge
3h 30m 25s logged
Was able to figure out how to get the rooms working, just need to clean up some things on both sides for this to be finished.
@jayeshsinghal
on
On Demand Judge
1h 53m 50s logged
Working on a room system, I was able to code the logic – still using the AI UI because I am simply too lazy right now, but I will get to it before ship.
You can invite people to debate you and the impartial AI scores you, is the concept so far.
Rooms need to be bugfixed, sitll working on that
@jayeshsinghal
on
On Demand Judge
1h 8m 19s logged
I was able to split the two into video and transcript, and one uses videos processed locally.
I want to see if i can make rooms and have people join and debate their friends… also redo UI
Loading more…