DATA
PREP
CHARTS
PLAN
EXP
RESULTS
Data Analysis
"how many companies?" · "plot ARR" →
Ask & explore
summarise, plot, segments, BI — answered in plain English the moment you load the file.
Data Science
"predict who hits $10M ARR" →
Train a model
forecast, predict, classify, optimize — OctOpus runs experiments and ships a model.pkl.
Data Engineering
"drop dupes · normalise arr" →
Clean & transform
clean, dedupe, normalise, ETL — every message becomes a transform step on your data.
Connect your data
CSV
CSV / Parquet
file upload
BigQuery
warehouse
Snowflake
warehouse
Databricks
warehouse
Redshift
warehouse
Postgres
database
MySQL
database
MongoDB
database
Amazon S3
object store
Google Sheets
SaaS
Notion
SaaS
ClickHouse
warehouse
yc_recent_batches.csv
| company_name | batch | industry | arr_usd | post_money_valuation_usd | mom_growth_pct | months_since_demo_day |
|---|---|---|---|---|---|---|
| YukonSec | S23 | Security | 541,153 | 42M | 14.2 | 12 |
| NovaOS | W23 | Dev Tools | 532,890 | 38M | 9.8 | 18 |
| SpireAPI | S22 | API Infra | 463,724 | 55M | 11.4 | 24 |
| EchoLabs | W24 | AI / ML | 421,224 | 120M | 22.5 | 6 |
| IrisPay | S22 | Fintech | 403,950 | 68M | 8.1 | 24 |
| HaloLabs | S23 | AI / ML | 253,504 | 72M | 17.8 | 12 |
| ZeroBio | W23 | Biotech | 235,961 | 28M | 6.2 | 18 |
| FluxSec | S24 | Security | 204,238 | 15M | 28.4 | 3 |
| ArcEngine | S24 | Dev Tools | 193,454 | 12M | 31.0 | 3 |
| KaiTech | W24 | AI / ML | 191,923 | 85M | 19.2 | 6 |
Data Engineering · live transforms
DE mode on
DROP
post_money_valuation_usd
remove column
FORMAT
founded
text → date · YYYY-MM
NEW COL
arr_per_employee
= arr_usd / employees
MERGE
+ funding_rounds.csv
left join on company_id
Result · 97 rows · 6 cols recipe saved
| company | founded | arr_musd | arr_per_employee NEW | last_round MERGED |
|---|---|---|---|---|
| YukonSec | 2023-04 | 0.54 | 31.8k | Seed |
| NovaOS | 2023-01 | 0.53 | 29.6k | Series A |
| SpireAPI | 2022-08 | 0.46 | 24.4k | Series A |
| EchoLabs | 2024-02 | 0.42 | 42.1k | Seed |
| IrisPay | 2022-06 | 0.40 | 18.4k | Series B |
Dashboard · YC cohort overview
auto-generated
$7.06M
Total ARR
97
Companies
14.8%
Median MoM growth
AI / ML
Top industry
ARR by industry
ARR trend · S22 → S24
Industry mix
MoM growth vs ARR
Research plan program.md
targetreached_10m_arr (binary derived from arr_usd)
taskclassification
metricROC-AUC, holdout-validated
splitgroup k-fold by batch (so later cohorts can't leak into training)
featuresindustry, founders, technical_founders, employees, last_round_usd, runway_months, mom_growth_pct, has_paying_customers, months_since_demo_day
experiments1) CatBoost · 2) LightGBM (Optuna 30 trials) · 3) TabPFN · 4) stacking ensemble
guardrailsleakage detection, holdout outside agent workspace, scrubbed env vars
Research loop
1
CatBoost · baseline
AUC 0.82
2
LightGBM · Optuna ×30
AUC 0.85
3
TabPFN · zero-shot
AUC 0.87
4
Stacking ensemble · 3 base learners
AUC 0.89
WINNER
Stacking ensemble · validated winner
0.89AUC · holdout
OctopusHi, I'm OctOpus, your AI Data Scientist. Drop a dataset and tell me your goal — I can train a model or answer questions about your data.
21:29
OctopusGot it. I can see your yc_recent_batches.csv dataset. Profiling it now…
21:30
drop duplicate rows, normalise arr to $M, and flag companies past $10M ARR
21:30
OctopusDE mode — applied 3 transform steps: removed 3 duplicates, normalised
arr_usd → arr_musd, and derived reached_10m_arr. Saved as a reusable recipe — your data is clean and model-ready.21:30
OctopusGOALWhat's your goal?
Skip
21:30
Data insight (Ask OctOpus)
21:30
OctopusInsights · FocusWhat should I inspect?
Skip
21:30
Summary
21:30
OctopusHere's what I see in this dataset:
It covers Y Combinator companies from recent batches — 17 columns capturing identity, funding, growth, and team composition.
Key financial signals available: arr_usd, last_round_usd, post_money_valuation_usd, runway_months. mom_growth_pct is the velocity signal.
Team data: founders, technical_founders, employees capture team composition.
Stage spread: has_paying_customers is a binary flag separating pre-revenue from revenue-generating companies.
Targets worth modeling: post_money_valuation_usd, runway_months, or a derived reached_10m_arr binary.
21:31
what's the total arr and the top companies?
21:31
OctopusAcross 100 companies, total ARR is $7,056,039. Top by ARR is YukonSec at $541,153 — there's a notable spread between leaders and the rest.
| company name | arr_usd |
|---|---|
| YukonSec | 541,153 |
| NovaOS | 532,890 |
| SpireAPI | 463,724 |
| EchoLabs | 421,224 |
| IrisPay | 403,950 |
+ 95 more rows
21:31
show me arr distribution across industries
21:31
OctopusARR distribution across industries
arr_usd ranges $0 → $541k, median $23.6k, mean $110.2k (right-skewed). AI/ML leads with $284k median.21:31
build a dashboard of the key metrics
21:31
OctopusDashboard ready. Built 4 KPI tiles and 4 charts — ARR by industry, ARR trend, industry mix, and growth vs ARR. Opened it in the workspace; export as PNG or embed anytime.
21:32
train a model that predicts $10M arr
21:32
OctopusPlan ready. Group-fold by batch so later cohorts can't leak. Open the
Plan tab to inspect, or type go.21:32
looks good, run it
21:32
OctopusRunning 4 experiments — CatBoost baseline, LightGBM with Optuna, TabPFN zero-shot, and a stacking ensemble of the strongest base learners.
21:33
OctopusDone. Best model: stacking ensemble at AUC 0.89 (holdout).
model.pkl ready in the Results tab.21:35
OctOpus is
idle
Ask anything about your data or model