SPARK Bot Academy · from the makers of SPARK ENGINE

Stop renting bots. Build your own.

A step-by-step course that takes you from an empty folder to a Solana trading bot that watches every Pump.fun graduation, follows your rules, paper-trades with real fees and runs 24/7 with a live dashboard. Every line of code is yours, and every line is explained.

Launch price €29. Goes to €49 after launch. No subscription, no profit fees, no hidden upsell.

python main.pypaper trading
19:04:51 GRADUATED7hWm…pump · following on DexScreener
19:04:52 SKIPPEDRUGX · pump-and-dump pattern ($412k right after graduating)
19:05:08 GRADUATEDFq2s…pump · following on DexScreener
19:05:09 BUYMOTH · 0.1 SOL · fresh graduate: $64k, 1s after graduation
19:05:41 SELLMOTH · 50% · take profit +22% (part)
19:05:57 GRADUATEDLp9d…pump · following on DexScreener
19:05:58 BUYKOI · 0.1 SOL · fresh graduate: $81k, 2s after graduation
19:06:12 SELLKOI · 100% · stop loss -16%
19:06:20 SELLMOTH · 50% · take profit +41%
19:06:35 SKIPPEDGOLDN · pump-and-dump pattern ($388k right after graduating)
18chapters, step by step
26project files, tested and ready
3systems: macOS, Windows, Linux
€0to build and paper-trade
What you build

A real bot with a real dashboard, not a script that prints prices

By the end you own a complete trading system that you understand line by line, and can change whenever you like.

●●●localhost:8080
The bot's dashboard: total result, open positions with live profit, finished trades and a feed of what the bot is doing.
The dashboard you build in chapter 10, updating every second. Shown here in demo mode: real results from the bot.
01

Hears every graduation

A live websocket to PumpPortal tells your bot the second a Pump.fun coin graduates to PumpSwap.

02

Decides with your rules

Entry filters, take-profits, stop loss, a crash guard and a pump-and-dump filter. All in one file you control.

03

Paper-trades honestly

Fills include pool fees and price impact, so paper results mean something before you risk a single SOL.

04

Goes live with brakes on

Real swaps through Jupiter only behind three switches, with a size cap and a daily loss limit.

05

Runs 24/7

On a €5 server, restarting itself after crashes, with Telegram alerts and the dashboard on your own domain.

How it fits together

Small files, one job each

Each box is one Python file of 50 to 250 lines. You write them in the order the data flows, one chapter at a time.

How the bot's parts connect PumpPortallive launches + graduations DexScreenerprice, market cap, pool size Jupiterreal swaps (live mode only) YOUR BOT · python main.py stream.pylistens prices.pychecks every 5 s brokers / live.pypaper or real fills bot.pyfollows coins,opens + closes trades strategy.pyyour rules server.pyweb + websocket Dashboardin your browser,or your phone trades.csvevery trade, logged
Dashed boxes are free outside services. Solid boxes are the files you write. Blue lines carry live data.
Curriculum

18 chapters, in the order you need them

Every chapter ends with a "done when" checklist, and every command is shown for macOS, Windows and Linux. About ten hours for a first-time builder.

Part 1

Foundations

  1. 00
    Start hereWhat you'll build, and what to expect
  2. 01
    How Solana memecoin trading worksWallets, graduation, pools, fees, rugs
  3. 02
    Your toolkitPython, VS Code, Git and the free accounts
Part 2

Build the bot

  1. 03
    Set up the projectFolder, virtual environment, libraries
  2. 04
    The Python bots useAsync, APIs and surviving errors
  3. 05
    Settings.env and config.py
  4. 06
    Live coinsstream.py: the PumpPortal websocket
  5. 07
    Pricesprices.py: DexScreener in batches
  6. 08
    Your rulesstrategy.py, and the lessons behind it
  7. 09
    Paper tradingbrokers.py and bot.py, the brain
  8. 10
    The dashboardWeb server, websocket and the page
  9. 11
    Put it togethermain.py and your first run
Part 3

Test, go live, run 24/7

  1. 12
    Paper-trade real coinsAnd read the results like a pro
  2. 13
    Going live safelyOwn wallet, Jupiter swaps, hard limits
  3. 14
    Running 24/7Home computer or VPS, systemd, Telegram alerts
  4. 15
    Your own domainDashboard on HTTPS with a password
  5. 16
    Improve the botRecord, replay, test on unseen data
  6. 17
    Security and troubleshootingChecklist, fixes and a glossary
Readable code

Short files you can actually read

No black box and no 3,000-line monster. Your rules live in one small file, and every decision the bot makes comes with a written reason in the log and the dashboard.

  • Three libraries in total: aiohttp, python-dotenv and solders
  • Paper trading by default; live trading is off until you switch it on
  • Demo mode with practice coins, so it runs on your first try
strategy.py
def entry_signal(coin, snap):
    """(buy?, why). Called on every price check."""
    age = snap["ts"] - coin["grad_ts"]
    if coin.get("pump_flag"):
        return False, "pump-and-dump pattern"
    if age > S.max_entry_age:
        return False, "too long after graduation"
    if snap["liq_sol"] <= 0:
        return False, "pool size unknown"
    if not S.min_mc_usd <= snap["mc_usd"] <= S.max_mc_usd:
        return False, "market cap outside range"
    return True, f"fresh graduate: {age:.0f}s after graduation"
Why it's different

Built from real testing, including what didn't work

Most bot courses show one great trade. This one is built on SPARK ENGINE's own research on thousands of Pump.fun coins, and it teaches you to test like that too.

−80 to −99%

Rugs jump right past stop losses

In 10-second recordings, rugged coins often fell 80–99% between two checks. That's why the course puts so much weight on which coins not to buy.

7–8×

Some coins are traps you can spot

Coins worth 7–8 times the normal graduation value seconds after graduating ended in a single crash every time SPARK ENGINE followed them long enough. Your bot learns to skip them.

≈ 3–4%

Fees decide more than you think

A round trip costs roughly 3–4% in fees and price impact. You'll see why small targets lose, and how to make paper results include every cost.

+20 / +40%

Take profit early

Coins that work often peak within minutes and fall back. The example strategy sells half at +20% and the rest at +40%, and the course shows you how to test your own targets.

Is it for you?

Made realistically, no get-rich-quick nonsense

A good fit if you…

  • want to understand and own the tool that trades for you
  • have never written a bot, or Python, but can follow steps carefully
  • are tired of paid bots that hide how they work and take a cut
  • like testing ideas with numbers before risking money

Not for you if you…

  • want a bot that "guarantees thousands a month"
  • want to skip paper trading and go straight to real SOL
  • are looking for front-running or other tactics that profit from other people's trades
  • can't afford to lose what you would trade with
No promises. Most memecoin traders, and most bots, lose money. This course teaches you to build, test and run a trading bot properly. It doesn't sell a money machine, and it isn't financial advice.
Get started

Your own bot, running tonight in demo mode

Buy once and keep everything: the full course as a page that works in any browser, plus the complete, tested project to compare your code with. Start with chapter 0 the moment your payment goes through.

Launch price
€29one-time · €49 after launch
  • The full course: 18 chapters with copy-ready code
  • The finished starter project (26 files)
  • Server files for running 24/7: systemd, macOS, Caddy
  • Paper trading, replay testing and Telegram alerts included
Get the course
One-time payment, VAT included. Secure checkout.
FAQ

Questions before you buy

Do I need to know how to code?

No. Chapter 4 covers exactly the Python a bot uses, with small programs you run yourself. You do need patience and the willingness to read an error message. Every file is given in full, so you're never stuck guessing.

Will the bot make me money?

The bot can make you money, given you have created and tested the right strategies. The course gives you an honest machine for testing ideas: paper trading with real fees, a replay tool and a way to score results. Whether a strategy works is something your own numbers show, over hundreds of trades.

What does it cost to run?

Nothing to build and paper-trade: every service used has a free option. Running 24/7 on a small server costs about €4–6 a month, and a domain for the dashboard about €10 a year. Both are optional.

Which computer do I need?

Any Mac, Windows PC or Linux machine that can run Python 3.11 or newer. The bot itself uses about 50 MB of memory.

Do I need to trade with real money?

No. The bot paper-trades by default and you can stay there as long as you like. Live trading needs three separate switches, its own wallet, and comes with a size cap and a daily loss limit.

How do I get the course after paying?

Right after checkout you land on your private download page with the course page (open it in any browser) and the project as a zip file. Bookmark it; you also get a receipt from Stripe by email.