Occam

$OCCAM has relaunched with a significant change in direction.

Live now, here is what you can expect:

✓ No bundle, launch is organic
✓ 90% of sell pressure removed from the hands of 20% supply regained from bad actors w/ zero output
✓ Chart optics cleaned
✓ Active, healthy & growth encouraging market-making to be running from day 1.
✓ Community able to claim their tokens 1:1 immediately

CA = 2sQ7wuUtRWNir3CEu9HWfLDSut4AszDrcZXLobzJpump

A Major Upgrade is Coming to Occam:

Release = Friday, 6th November.

We're doing the same again this week...

COUNTLESS updates everyday across last 7d:

- 22 Oct: A fresh start for Occam: honest financials, what's changing with $OCCAM, and the plan for profiles, data and AI.

- 23 Oct: Occam announcement on moonshot (bonding curve) algorithms are coming: small buys all day, chasing the 100x calls where 3/4 of pump.fun volume lives.

- 23 Oct: Occam announcement on now being discoverable by AI: Claude and ChatGPT can find and recommend us, with an MCP listing and Claude plugin ready to go.

- 24 Oct: Your AI can now build, check and backtest algorithms for you, with after-fee results and a check for lucky runs.

- 25 Oct: One peak everywhere: simulator, history and cards now match, and calls missed on 22–24 Oct are filled back in.

- 26 Oct: Tell your AI the result you want ("70% 2x every week") and it finds the rules, telling you plainly whether it's real or luck.

- 27 Oct: Gains now count only from your actual entry, and the simulator uses real entries, real prices and real fees.

- 27 Oct: Rug protection: bots refuse to buy when a pool drops under 20 SOL or the price falls 50%+ since the signal, with zero added delay.

- 28 Oct: Peaks are tracked at any time after the call, fake-price tokens are filtered out, and every card is rebuilt to match the simulator.

- 28 Oct: 41B transactions scanned, 2.95B trades measured, 70 TB of history rebuilt. Take the luck out of your backtests.

- 29 Oct: Signal Heatmap & Overfit Rating System

Automatically trade the market, with just your words.

Releases next week.

Time to play 🎲

Occam - Reviving the Business, Token & Excitement:

Updates, effectively immediately.

Let's dive into it, in this tweet we will cover:

> Financials & $OCCAM token
> Profiles
> Metrics & Data
> API & AI Compatibility
> Platform Volume & Market Expansion

There will be no beating around the bush here. No marketing speak, just setting the record straight. Truths & facts only.

💰 Financials & $OCCAM token:

We came far too close to having to close shop, too soon after launch.

Our incomings were minimal, our outgoings were very, very high. Approx $60k/mo at it's peak.

Here's what I think we did wrong:
> Took too long to launch, causing too much seed bloat as we required funding to build out (computational + staff - this was 16mo of consecutive outgoings, no incomings)
> Requiring us to launch at too high a MCAP.
> Consequently, the chart looked awful on day 1 as it came down to it's true-value price.
> The first impression anyone has of a crypto project, realistically - is their chart.
> The chart being ugly, put off potential users. If the chart is down only, there product won't be good. Skip.

The costs were crippling post launch. But, i'm sat on a product that has had near 7-figs of spend on, that truly works and is ultimately the best automated trading platform on the market.

There is a huge hole being burnt in my pocket. As much as I enjoy building it out, this is my living & full focus. It must be financially worth it to run.

So... what do we do?

We were sat approx 30/45 days ago with;

- $30k/mo computational costs
- $15k/mo staff costs

Bills overdue and hosting e.g AWS being taken down.

Since then;

1) Old bills have been covered by a small sale of personal equity, enough to cover outgoings. Not enough to hurt personal incentive.
2) Our aggressively high computational bills have been replaced by ones -95% lower, with no technical disadvantage.
3) Our metrics & data were previously provided by a partner. Internally, we have rewritten our own offering, with far more flexibility & customisation (more on this below...) with 85% lower costs compared to our previous billings.
4) Internal technical output (subjective) has vastly been improved

Compared to our previous peak costs of $60k/month, at this current day we operate at a -90% reduction in business running costs.

Giving us an incredibly sturdy & safe financial foundations, with a really healthy outlook & runway to continue building this out to see the heights it deserves to see.

🫂 Profiles:

As touched on before, AWS invoices were the reason for the downtime in losing access, this included user profile details (wallets, algorithms, autotraders)

This is inexcusable. Lesson learnt. Never again.

Profiles are now fully recovered, 1:1.

We have moved from AWS to Azure, with again, costs reduced by -90%.

No technical disadvantages.

You can sign back into your profiles as you could before, live now.

If you notice any differences/issues let us know ASAP and i'll resolve.

⚖️ Metrics & Data:

Again, as touched on before, our metrics were rewritten entirely.

The change from v1 to v2 before, was our provider improving the cadence (speed) of metrics, resulting in more metrics being able to be provided.

Now - there will be no v1 or v2.

Only one version, Occam's Metrics.

> It gives us much more flexibility.
> Users metric requests can be acted on immediately.
> Large cost reduction.

Most importantly... The ability to offer APIs & AI Compatibility which is the key to scaling, usage & volume.

In all of our tests, we saw incredibly high recall similarity from sim:real & precision rate (+99%) since our switchover on October 16th.

Between Oct-16th, expect 90%
In September, expect 80%

Here's what it means for you.

- Our own data: every metric on Occam is now computed by Occam from on-chain data with our own indexer. Nothing comes from a third-party data feed.

- One catalogue: Occam Metrics are the only metrics in the builder, simulator and marketplace. The old V2 metrics are retired. Algorithms built on them are stopped and marked retired, and their owners can rebuild them on Occam Metrics.

It's been a rollercoaster with the metrics & algorithms, but this'll be the last & final.

They are buildable with and trustworthy right now. Live.

What you are looking at, is our foundational version.

We will build up from the current offering constantly.

Spam us with metric requests, they'll be added.

🖥️ API & AI Compatibility:

It's becoming increasingly common that people are attempting to build automated / algorithmic trading bots themselves, using AI tooling.

The frequent limitations are:

> Lack of actionable & useful tradable data.
> The ease/platforming in executing what is built.

Here's our upcoming feature offering:

- Use Occam from Claude, ChatGPT or Grok
- Connect with one click, or use an API key
- Describe a strategy in your own words and get the rules built for you
- E.g, "Build me an algorithm that caught $KIMCHI below $100k MCAP but sticks to 5 tokens per day max"
- Mistakes get flagged before anything runs
- Backtest on the same data live trading uses
- See the numbers behind every token it picked
- Review, tweak and save your algorithms from your conversation
- Let AI create & publish for you directly onto the platform

"Build me an algorithm that has hit a 2x 70% of the time, consistently over the last month"
"How would you improve my currently published algorithm?"

A few key growth points:

> The hurdle on building an algorithm via the platform and learning what 100s of metrics each mean, is removed, but still available for those who wish to use this route.
> There should be significant influx of profitable algorithms published on the marketplace
> Our documentation is thorough & clear, with dozens of keys provided specifically to improve our 'SEO' via conversations with AI, drawing external market participants towards Occam of whom are looking to build an algorithmic trading strategy through Crypto. E.g, we'll be recommended as a solution to their problem/demand.

To be concise.

Both profitability & platform usage should increase substantially via a dramatic ease of use.

Release = Next week.

💹 Volume & Market Expansion

Tokens that haven't graduated yet, capture 3/4 of PumpFun's volume.

It is also where information & data moves quickest, giving the largest upside.

The large upside, that this market is particularly focussed towards.

To be direct - gamblers.

Admittedly, an audience I think we've been previously unappealing to with our high-brow offering.

We will be releasing full compatibility to newly launched tokens, right from the first transaction.

Features:

- Catch tokens the moment they launch, before they graduate
- 56 new metrics built specifically for brand-new launches
- Switch on new launches per algorithm with one toggle
- Backtest new-launch strategies like any other
- New-launch calls are tagged and link straight to pump.fun
- No double calls when a token graduates
- Forced safety nets on +99% chance rugs, e.g dead or stalled launches skipped
- Filter trading pre-graduation tokens out on an algorithm you run or build

Why didn't we do this before?

Truthfully, the scale of it was unbearable for our product offering.

100s of simulations on 40,000 daily tokens was too much load.

What we've built now, makes it fully possible.

Both with metrics being built ourselves but AI compatibility allowing the scale to be digestible.

Once again, opening us up to substantially more volume & user opportunity.

Release = Next week.

--------

- Signals, simulator and site now run on Occam's own on-chain data
- One clean metric list in the builder: Occam Metrics
- Health tags show which metrics you can trust
- Bot-laddered tokens kept out of signals
- Risky bundled launches blocked from signals
- One call per token per algorithm every 24 hours
- No more burst of repeat calls at midnight
- Age rules count from launch and catch fresh graduates
- Cleaner builder: clearer labels, finer price precision, no dead entries
- Simulator signals now match live trading
- Much faster simulator, and 30-day backtests now finish
- Simulating a saved algorithm is faster and closer to live
- Simulator results shown after fees
- Every figure labelled "before fees" or "after fees"
- 1D, 7D and 30D views show only tokens launched in that window
- Realistic all-time highs on algorithm pages
- Fixed marketplace stats: tokens per day, averages, top market caps
- Leaderboard Average X counts each signal once
- Private algorithms stay private, simulations included
- Wallets refuse any transaction that would move funds off your wallet
- Trading, deposits and withdrawals back open
- Every pre-migration account can sign in again

———-

In a matter of 45 days, Occam has gone from a place of conversations happening in nearly needing to shutdown - to a financially secure, fully functional platform with all use restored & plenty of exciting features in the immediate release pipeline.

Albeit a low floor, we have held substantially well for the last 45 days considering the entire market has been in downfall and the product has been unusable.

We are in good stead, experienced, funded & low outgoings with a ton of rollouts ready to push.

Absolute focus from here - is on marketing & platform usage.

From there, follows token price activity.

No plans in stopping, time to violently reprice this business.

Thank you for your patience, we've made it through the other end of winter.