CEO & Founder at Fuelfinance / make finance easy for founders

I never planned to become a founder.

I spent most of my life as a CFO — building models, analyzing numbers, and helping businesses grow. I loved it. Still do.

But in 2019, something shifted. I built a simple Google Sheets MVP to help my first customer (Petcube, YC’18) understand their finances better. Five months later, they were profitable. A year later, that client became my co-founder.

That’s how Fuel started.
Today, Fuel is a financial management platform paired with a real CFO. 

We manage over $500M in customer P&Ls. 
We’ve grown from a no-code MVP to a full web platform used by hundreds of businesses.

Along the way:

→ Partnered with HubSpot, Carta and Brex
→ Helped 100+ agencies and SMBs become profitable: Petcube, ColdIQ, Oceans, Hampton & more

Outside of Fuel, I’m:

→ A cross-continental swimmer (4.04 miles from Europe to Asia in open water)
→ Dog mom to Haku, my husky (and Fuel’s unofficial Chief Food Officer)
→ Into skateboarding, wakesurfing, snowboarding (anything that reminds me to feel fear and move through it)
→ Exploring longevity — 1.5 years alcohol-free, lifting weights, and learning to cook for myself

My mission is simple: make finance easy (and sexy) for founders.

If you wanna have a founder-to-founder chat, shoot me a DM.

Alyona Mysko

Y Combinator just put "AI-native agencies" on their official Requests for Startups list.

YC (the incubator that backed Airbnb, Stripe, Dropbox) is now actively asking founders to build agencies.

I've worked as a CFO with 100+ agencies. 
And I never thought I'd see this day.

👇 Here's why this is historic.

For decades, VCs looked at agencies and said: "Not scalable. Low margins. Too people-dependent. Next."

Agencies were the businesses that couldn't get funded. The ones told to "pivot to SaaS" if they wanted investment. Revenue scaled linearly with headcount - and that was a dealbreaker.

Now YC is saying the exact opposite. They're saying agencies of the future will have software margins.

What changed? AI.

From YC's RFS: instead of selling software to help customers do the work, you use the software yourself and sell the finished product.

A design firm that delivers custom work before the contract is signed. An ad agency that creates video ads without a physical shoot. A law firm that writes legal docs in minutes, not weeks.

Same output. Fraction of the cost.

Agencies are our core customers at Fuel. I've watched their P&Ls for years. The ones adopting AI right now see their numbers transforming.

Gross margins moving from 40-50% to 70-80%.
Delivery timelines shrinking 3-5x.
Revenue per employee that looks like SaaS.

→ The old formula: more revenue = more people = more overhead = thin margins. 

→ The new formula: more revenue = better AI workflows = same team = software margins.

If you run an agency and you haven't rethought your delivery model yet - the world's most famous startup accelerator just told you: your business model is the future.

Don't waste that moment.

Alyona Mysko 

Founder of Fuelfinance | building the future of finance for SMBs

Being a woman founder is hard. 
But being a woman CFO is a whole different game.

I've been on both sides.

I spent over 10 years in finance - auditor, analyst, then CFO. Then I became a founder.

The founder path gets all the attention. The headlines. The "women in leadership" features.

But no one talks about the woman CFO.

As a woman founder, it's brutal - don't get me wrong. Women founders still raise only ~2% of all venture capital. You get questioned more. Tested constantly.

But you have one thing: freedom. 

You set the vision. You set the culture. You get to make a mistake, own it, learn from it, and move on. The company is yours. The story is yours.

As a woman CFO, you carry the same weight (sometimes more) but none of the narrative.

You're the one who tells the CEO the runway is 4 months, not 8. You're the one saying "we can't afford this" while everyone else is saying "let's go." 

You are the person who delivers the hard truth - and then gets blamed for being "too conservative" or "not a team player."

And you do all of this while navigating boardrooms where you're often the only woman at the table.

The CFO doesn't get to set the rules. She has to survive someone else's rules - and still protect the company.

I went from the CFO seat to the founder seat. I know exactly how different those pressures feel.

And I have enormous respect for the women who show up in either role. Every. Single. Day.

I'm glad I know amazing women on both sides.

→ Amy Hood - Microsoft's first female CFO, overseeing an $80B AI investment year.
→ Colette Kress - led Nvidia through its rise to one of the world's most valuable companies.
→ Susan Li - became Meta's CFO at 39 and is now managing $64-72B in capex.
→ Kate Bueker - running finance at HubSpot.
→ Margarita Chavez - CFO of Endeavor, the org behind Glovo, ElevenLabs, and Brex (had her on our podcast - total operator energy).

Tag your favorite woman founder or CFO in the comments. 
Let's make sure they know we see them 💚

Here's what happened in AI in February that you NEED to know as a CFO or finance leader:

Before I go to the news, it's fair to say that I've been pretty sceptical about AI for finance, but February 2026 genuinely changed how I think about it.

Here's what happened:

1/ Anthropic released Claude Opus 4.6 - their most powerful model yet.

→ If you've been using Claude for finance work, the jump is VERY noticeable. If you haven't tried it yet - this is the model that makes it worth testing. The 1M context window means you can feed it an entire annual report or a full set of financials and actually get useful analysis back.

2/ Claude now lives inside Excel and PowerPoint - and they talk to each other.

Anthropic upgraded Claude in Excel (pivot tables, charts, conditional formatting via natural language), launched Claude in PowerPoint, then connected the two. You can now run a financial analysis in Excel and have Claude build a board presentation from the results.

→ Try it on your next monthly reporting cycle. Start with a variance analysis deck. Measure the time saved.

3/ Anthropic launched finance-specific Cowork plugins with real market data.

Connectors to FactSet, S&P Capital IQ, MSCI, and LSEG - inside Claude. Plus five finance plugin templates: Financial Analysis, Investment Banking, Equity Research, Private Equity, Wealth Management. Your analysts can pull live data, run analysis, and generate outputs without switching between six terminals.

→ Check if this overlaps with tools you're already paying for. Potential cost consolidation.

4/ More finance tools are adding Claude MCPs directly inside their products. Rillet's one is the most recent one I've seen.

→ Check what your current CFO tech stack is shipping. If your accounting, FP&A, or ERP provider added an MCP, you may already have AI capabilities you're not using.

5/ Salesforce, Workday, and ServiceNow all shifted how they price and deliver AI.

Every major software vendor is moving from "AI-assisted" to "AI-autonomous" and from per-seat to consumption-based pricing. Your software budget is becoming a variable cost line.

→ Audit your current enterprise contracts. Model the pricing shift impact before renewal. And have a real conversation with your team about which roles are being augmented vs. replaced - because your vendors already are.


This is just what caught my eye as a CFO and someone building in this space.
What are you seeing on your end?

Here’s what AI won’t replace in finance.

A CFO showed me a framework last week that made me exhale for the first time in months. It was about "post-AI human work."

The idea is simple: every job gets split into three layers.

Transactional tasks → gone. AI handles these. 
Knowledge and expertise → shared between humans and AI. 
People-oriented work → stays human. Fully.

She said: "I find comfort in that. Like okay, now I can see how my job will be divided." 

And I realized - that's exactly what I've been missing.

I've been so deep in the AI world that I forgot to zoom out and ask: what's actually left for us? 

So I thought about it through the lens of a CFO. 

🟩 What AI is already replacing in finance:

→ Data entry 
→ Transaction categorization 
→ Report formatting 
→ Reconciliation matching 
→ Basic variance calculations

🟩 What AI is starting to share with us:

→ Forecasting 
→ Anomaly detection 
→ Scenario modeling 
→ Board deck drafts 
→ Contract analysis

🟩 What stays human:

→ Explaining to the CEO why the numbers matter 
→ Telling the board, “we need to change strategy” 
→ Taking responsibility (AI can’t do that) 
→ Building trust with investors 
→ Communication and people management - everything that still requires human interaction

The transactional layer is disappearing fast. Honestly, it should.

Nobody became a CFO to copy-paste bank data into a spreadsheet.

But the people layer is getting more valuable. The CFOs who will thrive aren't the ones who learn every AI tool.

They're the ones who get really good at the work AI can't do - and let AI handle everything else.

I spent a decade as a CFO doing both. If I could go back, I'd spend 80% less time on spreadsheets and 80% more time in the room making decisions.

Thank you Margarita for this insight!

If you're in finance and you can't build a basic AI agent by end of 2026, you're the new person who can't use Excel.

Here's where to start so you're not behind:

1/ Pick your most repetitive weekly task. The one you hate. The one that's mostly copying data from one place and summarizing it in another.

2/ Open Claude (you can learn a lot just from tutorials on their website). Describe the task. Ask it to help you automate it. You'll be surprised how far you get in an hour.

3/ If it works even 70% - keep refining. You just built your first automation.

4/ Try Claude in Excel if you've been building spreadsheets manually. Game changer.

5/ Try connecting Claude to PowerPoint. Board decks in minutes, not hours.

6/ Explore AI-native finance tools that replace your legacy systems. The ROI will surprise you.

7/ Check whether your current vendors already shipped AI features. You might save hours just by changing how you use the software you're already paying for.


You don't need a course. 
There's no course that can keep up with AI right now - it's changing every day.

Honestly, I feel a bit overwhelmed myself with everything happening in AI.
But the only way is to grab a surfboard and ride the wave.

So tell me - where are you?

a) Riding the wave 🏄‍♀️ 
b) Watching the wave from the beach 
c) Didn't know there was a wave 
d) The wave knocked me over and I'm underwater 
e) I AM the wave (to the AI CFOs out there)

Wow. Just weeks after Claude's huge Excel release, OpenAI dropped ChatGPT for Excel. 

As someone who tested both - this is getting exciting.

👇 Here's what's happening:

Anthropic went first. Claude in Excel started as a waitlist of 1,000 users back in late 2025. I tested it early and shared my honest review - the data cleanup alone saves hours every week.

For basic modeling, it builds solid structures with correct formulas. For advanced modeling, you still need a human brain. Because in finance, 99% correct is still wrong.

But what Claude did was something bigger than a feature launch. It proved that AI belongs inside the spreadsheet, not next to it.

Now OpenAI followed. ChatGPT for Excel, powered by GPT-5.4, launched March 5.

It does many of the same things: build models, run scenarios, explain formulas, update workbooks using plain language. They added financial data integrations — Moody's, FactSet, MSCI, Dow Jones Factiva - very similar to what Anthropic did with S&P Capital IQ, LSEG, and Moody's.

The investment banking benchmark numbers are impressive: 87.3% on GPT-5.4 vs. 43.7% on GPT-5.

But here's what I actually care about as someone who spent 10+ years in finance and now builds an FP&A product:

🟩 The competition isn't about who has the better model anymore. It's about who understands how finance people actually work.

A few observations after testing both:

→ Claude understands workbook structure better. Cross-tab tracing, cell-level citations, formula logic - this is where it shines. When you inherit someone else's messy 15-tab model (every CFO knows this pain), Claude is genuinely helpful.

→ ChatGPT has broader reach. Available on Plus plans. More financial data integrations at launch. Google Sheets coming soon. OpenAI is playing the distribution game - and they're good at it.

Neither replaces a finance professional. Both still hard-code values they shouldn't. Both miss assumptions a good analyst would catch. Both need a human reviewing the output.

And that's fine. That's where we are.

6 months ago, the question was "should we use AI in our financial workflows?" Today, the question is "which AI is embedded in your Excel?"

That shift happened fast.

🟩 For founders: if you're still doing your financial planning manually, you're already behind. These tools won't replace your CFO or your FP&A team. But they will make the good ones 3x faster.

🟩 For CFOs and finance leaders: test both. Form your own opinion. Don't trust the benchmarks - trust your own models and your own data.

This race is just getting started.

If you’ve worked in finance long enough, you know this:
The traditional finance org is already dead.

For our 2026 CFO AI Stack Report (link’s in comments), we asked founders and finance leaders one question:

How is AI actually changing your category?

Here’s what we’re seeing:

→ FP&A: Copilots make financial analysis available in plain English - even for marketing and GTM teams.
→ ERP: Month-end close is shrinking. Reconciliation is becoming background work, not a 5–14 day marathon.
→ Billing & RevRec: AI catches issues earlier. Humans still approve.
→ Tax: Automatic nexus tracking and notice parsing.
→ AP: Most routine processing gets automated. Teams shift to managing cash and risk instead of chasing invoices.
→ Procurement: AI surfaces renewals, anomalies, and negotiation prep before someone remembers to ask.

The big shift:

Humans are moving from execution → to judgment.
From preparing numbers → to interpreting them.
From reporting history → to shaping decisions.

Scroll through the carousel for direct insights from founders building these tools.

And if you see something differently (or have a use case we missed) drop it in the comments.

- 

Huge thank you to Jonathan Sanders, Sam Ross, Anshul Shah, Russell Lester, Ali Hussain, and their teams at Light, Numeral, Brex, Tropic, and Tabs for helping shape this report.

Link in comments ⬇️

AI CFO stack report

A year ago, Dirk Sahlmer commented on one of my LinkedIn posts about M&A.

It turned into a long conversation about how founders actually experience exits vs how exits are usually talked about online.

Multiples, headlines, big numbers - but very little about what really happens behind the scenes.

Dirk has advised on hundreds of transactions at FE International and saas.group, working with founders through valuation, diligence, negotiations, and the messy middle part most people only discover once they’re already in it.

We started exchanging notes from our sides:

– what founders think buyers care about
– what buyers actually look for
– how financial preparation changes outcomes years before a deal happens

That conversation eventually turned into a full valuation guide for founders ( ⬇️ ).

And now we’re turning it into a LIVE session.

On Feb 25th, we’re hosting a webinar together with Arman Taheri, who recently went through a strategic investment process that valued TalentPop at nine figures.

We’ll talk about:

– when a business becomes truly sellable
– how valuation really works beyond online multiples
– what buyers look for during diligence
– common mistakes founders make during M&A
– and how to prepare your finances 2-3 years before an exit or investment

If you’re building a profitable business (or even thinking about an exit/investment one day) this is the kind of conversation I wish more founders heard earlier.

RSVP in comments ⬇️

54% of CFOs plan to integrate AI agents into finance in 2026.

Over the past year, most conversations I’ve had with CFOs shifted from “is AI actually useful for finance?” to “where do we start delegating work to it?”

2024 was skepticism.
2025 became experimentation.
2026 looks like implementation.

Finance teams are overloaded with structured, repeatable work - updating plans, explaining variances, preparing reports, and answering the same questions every month.

AI agents don’t replace finance teams.

They remove the manual layer around decision-making.

What I’m seeing is finance slowly moving from:

producing reports → reviewing insights
collecting data → interpreting it
reacting monthly → operating in real time

That shift is a big reason why we’re building Fuel - AI agents for finance teams that help automate analysis and make financial visibility accessible beyond the finance department.

This chart is a preview from our 2026 CFO AI Stack Report (mid-market edition), where we break down how AI is changing the entire CFO stack.

(Download the full report via the link in the comments ⬇️ )

Curious how others see it:

Where do you personally see AI agents having the biggest impact in finance first - FP&A, accounting, or somewhere else?

Selling your business can cost $350K in legal fees alone. Here's what else no one tells you about exits 👇

I sat down with an M&A advisor (20+ exits) and a founder who just closed a 9-figure strategic deal. Full recording is in the comments.

Here's the TL;DR.

On valuation:

→ 6-7 figure revenue businesses: 3x-15x revenue multiples depending on growth, retention, risk. 
→ Profitable service/tech businesses ~$10M revenue: roughly 3-6x EBITDA. 
→ AI-enabled businesses command higher multiples. Non-AI businesses are seeing compression.

On knowing your business is sellable:

→ Have real exit intent. Not "maybe exploring." This process is 12-18 months and a second full-time job. 
→ Get shareholder alignment BEFORE you start. Deals die because a small angel or ex co-founder renegotiates at Letter of Intent (LOI) stage. 
→ Focus on your business, not the macro. Your fundamentals matter more than public markets.

On financials:

→ Rule of 40 still matters. Can't grow fast? Push profitability. Dirk has seen good exits with flat revenue but 30%+ EBITDA margins. 
→ Get an external QoE done. Don't DIY it - buyers will forensically audit you. The sell-side QoE pays for itself 10x. 
→ #1 reason deals get retraded after LOI: inconsistencies in the numbers.

On the auction process:

→ Never sell without an advisor. Dirk's team has more than doubled initial offers by running a proper auction. 
→ Before LOI - you have leverage. After LOI - it shifts to the buyer. 
→ Evaluate buyers by effort, not just the number on paper.

On costs:

→ From experience: you can have up to ~$350K in legal fees alone. It's an investment, not an expense. 
→ Consider tiered fee structures where advisors earn more at higher valuations.  Aligns incentives.

On taxes (the part founders forget):

→ C-Corp vs S-Corp changes everything. Look into QSBS early. 
→ Earnouts can be taxed as income, not capital gains. 
→ California is not your friend here.

On when to sell:

→ There's no formula. It's deeply personal. 
→ If you're running out of steam - buyers will sense it too. 
→ Don't wait for "the perfect time." Sometimes the right time is now.

Arman's advice: Be authentic. Make sure everyone in the deal wins. 
Dirk's advice: Prepare earlier than you think. If you're 1-2 years out, start talking to advisors now.

Full recording in the comments 👇 
If you're a founder thinking about this in the next 1-3 years, this one's worth the watch.

3 big changes happening across finance teams:

1️⃣ Legacy tools are being replaced by AI-native systems.

2️⃣ A huge part of finance work will be delegated to agents - emails, Slack inputs, plan updates. Humans review and approve, not execute.

3️⃣ Finance teams are redesigning workflows before hiring. Headcount follows automation, not the other way around.

In ERP (the category that received the most attention and funding last year), AI is making real-time close possible.

In FP&A, AI copilots make financial analysis accessible in plain English - even for non-finance teams.

In Tax, AI now enables automatic monitoring of state thresholds and government notices (a game changer).

__

This is a preview from our 2026 CFO AI Stack Report (mid-market edition).
15+ pages.

50+ tools mapped across ERP, FP&A, billing, rev rec, AP, tax, procurement, close, and more.

Built from conversations with 100+ CFOs and founders.

If you’re running a mid-market finance team, this will save you weeks of research.

Download link in the comments 👇
And tell me, what’s the biggest AI use case in your finance team today?

(get the full report in the comments) 

2 months in, and 2026 has already delivered some of the largest fintech exits in history.

If you thought M&A would slow down - it didn’t.

Here are a few that stood out:

1/ Capital One × Brex — $5.15B

Capital One agreed to acquire Brex in what’s being called one of the largest bank-fintech deals ever. Yes, the valuation is lower than 2021 highs - but $5B+ outcomes are still real outcomes.

2/ Hg × OneStream — $6.4B

Hg is acquiring OneStream, a finance management platform, in an all-cash deal. Private equity doubling down on finance software tells you one thing: CFO tech is strategic, not experimental.

3/ Deutsche Börse × Allfunds — €5.3B

A stock exchange operator acquiring a wealth-tech platform.
Translation: distribution + data + platforms are becoming core assets in capital markets.

4/ US Bancorp × BTIG — up to $1B

Institutional trading and investment banking capabilities moving further under large banking umbrellas. Traditional finance continues absorbing niche specialists.

5/ Airwallex × Paynuri (South Korea)

Expanding licenses and geographic coverage. Cross-border payments are still a land grab.

6/ Visa × Prisma Medios de Pago & Newpay (Argentina)

Visa is strengthening its position in Latin America by acquiring Prisma Medios de Pago and Newpay. This is about owning local rails and distribution in high-growth markets - not just issuing cards, but controlling the ecosystem.

—

And that’s your early 2026 exits wrap.

But here’s the part most founders miss:

Exits don’t start when someone sends you a term sheet.
They start years earlier - in how you build, structure, and present your business.

On Feb 25th, I’m hosting a live session with:

– Dirk Sahlmer, Partner at FE International (he’s advised on hundreds of exits)
– Arman Taheri, who recently went through a 9-figure strategic investment process

We’ll break down:

- when a business becomes truly sellable
- how valuation actually works
- what buyers look for in diligence
- how to prepare your finances 2-3 years before an exit

If 2026 is already this active, it’s worth understanding how the game actually works.

Link in comments.

Our customer is a founder + the only person running finance at a $12M logistics company.

Payroll questions: him. Benefits issues: him. Weekly reports for leadership: him.
Quarterly board decks: also him - searching his computer at 9pm for last quarter's PowerPoint and the Excel he copy-pasted the numbers from.

Every Monday morning he'd pull data from his systems, manually type it into a spreadsheet, build a chart, and present it to the team. If something changed over the weekend, he'd go back week by week to find it.

Sometimes all the way back to September.

When I first looked at what he built, I was impressed. He knows his numbers cold. Tracks revenue mix, margins by product line, individual rep performance.

The problem was never the thinking. It was the manual work eating ALL his time.

This is exactly what we built Fuelfinance for.

We connected his systems to Fuel. Now he hits refresh and everything updates automatically - including that load from September.

Once the data was clean, he could clearly see that one type of shipment was generating 20% margins while another was sitting at 8%. Now he has the numbers to actually shift his sales team's focus toward the more profitable work.

But honestly the biggest win isn't even the hours saved each month. It's knowing that finance is just running. Without him having to hold it all in his head while someone's calling about benefits and the leadership meeting starts in 20 minutes.

Three months in, he rated Fuel 9.5 out of 10.
When I asked what would make it a 10, he said:
"Start telling me what I should be looking at."

We're working on that 10.

AI suggestions, a Copilot he can talk to about his numbers anytime, AI-powered summaries and more. We're also building in deep customer context - so the AI actually knows his business, his margins, his team, his goals. Not generic answers, but recommendations grounded in his own data.

If you're a founder running finance alone - DM me.
I'd love to show you what Fuel can do.

My friend Arman grew TalentPop to a 2500 team member org in <6 years.
Here’s everything he wishes every founder knew before planning their investment or exit strategy:

1/ Most founders have a job and don’t have an investable or sellable business.

You know your revenue and growth - but buyers look at risk, dependency, structure, and predictability. What you think your multiple is often isn’t reality.

2/ The fundraising takes longer (and is heavier) than you expect.

Months of diligence, negotiations, and valuation discussions happen while you still need to grow the business. It’s not for the faint of heart.

3/ You can’t do it alone.

You need the right people around you - investment bankers, legal, tax, finance leadership, and business advisors. Trying to run the process yourself often means leaving money on the table.

4/ Investors or buyers need to win too.

An investor or buyer is a partner. If there isn’t upside for the opposing party, the deal doesn’t work. The best outcomes come from give-and-take, not maximizing every term.

5/ The hardest questions start after the deal closes.

Tax planning, capital allocation, and how to preserve what you’ve built are things founders rarely think about early enough.

_

We’re turning this into a live conversation on Feb 25th, where Arman will share the full story together with Dirk Sahlmer (FE International) and myself - including what surprised him most during the process and what founders usually get wrong.

If you’re building a profitable business and want optionality in the future, this will be a very practical session.

Link to register in the comments 👇

AI made finance accessible for every team.

- marketing can ask: “What happens if we increase CAC by 20%?”
- GTM can check payback periods before launching a new channel.
- product can finally see how usage changes revenue - without waiting for a report.

This is new.

For years, financial analysis lived inside finance teams and spreadsheets.
You needed models, formulas, and someone who knew where the numbers actually lived.

Now copilots changed that.

Non-finance teams can ask questions in plain English and get answers in seconds.

And honestly - this is a good thing.

Finance shouldn’t be a bottleneck.
It should be infrastructure for better decisions.

What I’m seeing already:

– marketing teams testing scenarios before spending budget
– GTM teams understanding unit economics earlier
– founders asking better questions because data is accessible
– finance teams spending less time building reports, more time reviewing decisions

At Fuel, this is exactly what we’re building toward - financial analysis that’s accessible beyond finance, but still grounded in real data and finance logic.

Curious how this looks inside your teams:

Marketing, GTM - where do you actually run financial analysis today?
Claude? ChatGPT? AI copilots inside finance tools?

Or still going through finance for every answer?

In the middle of the biggest AI wave, 82% of finance pros are still emotionally attached to Excel.

Half said they'd quit a job that didn't let them use it.

A recent survey of 212 finance pros (CFOs, accountants, analysts) found:

→ 45% say they "love" Excel. Another 43% say "love-hate." 
→ 99% spend at least 3 hours a day in it. 
→ Gen Z and millennials: 83% are in Excel 5+ hours a day. 
→ 84% of CFOs say Excel will be equally or more important in 10 years.

All while we already have AI that can build financial models, automate reconciliations, and draft board decks.

Here's what I think:

Excel isn't surviving despite AI. It's surviving because of what AI hasn't solved yet: the feeling of being in control of your numbers. Claude launched an Excel add-in. ChatGPT did too. If you can't beat the spreadsheet, join it.

But honestly, the Excel vs AI debate is already the wrong question. 

The real shift happening right now is agents.

AI that doesn't just live inside your spreadsheet but actually does the work - reconciliations, forecasts, month-end close, board prep. Where it runs (Excel, Google Sheets, an ERP) won't matter much longer.

What matters is how much manual work gets removed. And we're talking 100x, not 10%.

That's the question finance teams should be asking: not "will Excel survive" but "how much of what my team does manually today should still be manual in two years."

The tool is not the point. The work that disappears is.

Yurii Filipchuk 🇺🇦 -- e/acc

Founder @ Stealth AI |TechStars AI Startup Weekend SF

I build Principle, a simulation engine that stress-tests business decisions before you commit capital.

I also run partnerships and sponsorships at Techstars Startup Weekend SF and Catalyst Bay. We're a community of 10,000 founders doing IRL events in San Francisco.

Chaotic Good is my operating system. Move fast, challenge conventional thinking, put powerful tools in more hands. I believe that's how you actually make the world better. Big believer in pay-it-forward culture. If you have an idea on how we can help you or you want to help us, reach out.

The LLMs you choose say something about you. I'm team Claude for work. NotebookLM and Gemini for research. ChatGPT and Groq for personal life questions.

Previously CEO of Party.Space. Spotify was my first client, then Google, Zapier, Grammarly, UC Berkeley. Raised $2M from top European VCs.

I advise early-stage founders on fundraising and GTM. Been on both sides of the table. Invested through CYFRD: 15 deals, 2 exits.

San Francisco. Born in Kyiv.

Community Lead: Vibes & Sponsorships | Pitching Mentor

At Techstars Startup Weekend SF AI, we're hosting a fast-paced, 3-day event where tech enthusiasts come together to bring AI startup ideas to life. My role is to create an energetic and welcoming atmosphere, build connections with important sponsors, and guide participants in making their pitches clear and impactful.

Google DeepMind just ran their first Startup Session.

Got the invite from Matt Ridenour - thanks for putting this together and MCing the whole thing.

Direct access to the teams building the models. That's rare.

What they showed:

1/ Austin Sendek on materials science - his team's GNoME project discovered 2.2 million new crystal structures. That's roughly 800 years of materials knowledge generated by AI. Wild number.

2/ Genie 3 - a generative world model trained on internet videos. You describe an environment, it builds a playable 3D world in real time at 720p. Already shipping as Project Genie for AI Ultra subscribers.

3/ Anshul Ramachandran on Antigravity - DeepMind's agent-first IDE. An AI Editor, Agent Manager and Agent-Controlled Browser working together. Task-level coding, not line-by-line.

4/ Nano Banana - their image generation model that pulled 10M+ new Gemini users and 200M+ edits in the first weeks. Now powering the whole Genie world-building pipeline.

And Paige Bailey generated Hindu electro pop live on stage with Lyria. Totally awesome!

Great seeing old friends like Naji Barnes-McFarlane!

Google've been always supporting startups. Good to see them opening the lab doors.

Yurii Filipchuk

AI Simulation Platform for Business Decisions | Chaotic Good 🇺🇦

The state of GenAI video no longer surprises me🤖

So we started making our own. Episode 1 of "LockedIn" is out. Who else gets frustrated receiving AWS/GCP bills?

Where should we channel the creative energy next -- something more satirical or slightly educational?

P.S. Those worms are adorable and they deserve their own show. Someone talk us out of it.

We act like AI showed up yesterday.

1951 - Christopher Strachey wrote a checkers program on the Ferranti Mark 1. A room-sized machine playing board games. That was the first AI.

1966 - MIT built ELIZA. A chatbot pretending to be a psychotherapist. People got emotionally attached to it. 58 years before ChatGPT.

1972 - MYCIN diagnosed blood infections better than most doctors. An AI system. In the seventies.

By the 1980s, companies were spending millions on expert systems that mimicked specialist knowledge. Then the hype died. Funding dried up. They called it the "AI winter."

We've been through this cycle before.

When was the first time you used a product and thought "wait, is this AI?"

early ai cases 

Dropping a new episode of LockedIn. Inspired by a real story this time!

Hamudi Naanaa and Volodymyr Panchenko hit Anthropic's Tier 4 token limit mid-hypergrowth. 4 users to 1,400 in 3 days. 2b tokens a day! 

Their words: "A two-day response cycle doesn't match the speed we're operating at -- for us, it's 1–2 hours."

They switched to Gemini overnight. A constraint forced it.

What's the ceiling your team hasn't hit yet -- but will?

taplio

The Battle

Alyona Mysko

Yurii Filipchuk 🇺🇦 -- e/acc

Followers

Posts

Engagements

Likes

Performance ScoreAlyona Mysko's LinkedIn Performance

Performance ScoreYurii Filipchuk 🇺🇦 -- e/acc's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

Alyona Mysko

Yurii Filipchuk 🇺🇦 -- e/acc

Followers

Posts

Engagements

Likes

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Alyona Mysko's LinkedIn Performance

Performance ScoreYurii Filipchuk 🇺🇦 -- e/acc's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance ScoreAlyona Mysko's LinkedIn Performance