AI-native software for finance teams that close faster, cleaner, and without added headcount.
Amsterdam
Joined December 2024
- Tweets70
- Following50
- Followers167
- Likes74
Pinned Tweet
"Stacks felt like it was built for the way we actually want to work." - Ida Salmen, Cost Accounting Manager at @epidemicsound
The full story of how their finance team rebuilt their close 👇
Excited to share our keynote for the Stacks Summit 2026: Sir @Jeremy_Hunt, former UK Chancellor, making the case for AI as Britain's next growth engine.
Every attendee will get a free signed copy of his new book, Can We Be Rich Again?, on 29 Sept in London.
Link in reply.
@BoucherNicolas, building finance workflows live and calling build versus buy
@AnthropicAI, the team behind @claudeai, as event partner and on stage
@JDWWatkins of @AltimaSearch on what AI does to finance talent
A Big Four audit partner on what it takes to sign off
Leading CFOs on what they have learned
Today we shipped an agent that reconciles and resolves your VAT.
A deterministic three-way match across your GL, VAT return and the tax authority's figures, before you file. Every difference explained in plain English, traced to source.
Recs in minutes, not days.
Reassigning checklist tasks one by one is the kind of small friction that quietly slows down a close. People go on leave, ownership shifts, due dates move.
Bulk actions fix that. Select the tasks, an action bar appears at the bottom of the screen, reassign preparer or reviewer, shift due dates, or delete in one go. Conflicts stay surfaced so nothing slips.
We just shipped the biggest Flux update since launch.
Until today, Flux only did MoM variance. It now runs four comparison modes, with AI explanations across all of them:
→ Monthly
→ Quarterly (any 3-month block)
→ Yearly (any 12-month block)
→ Custom (incl. fiscal YTD)
Use flux analysis for board decks, auditors, group reporting and more: all in one tool.
Stacks retweeted
Shipping AI is one thing. Shipping AI that actually works in production is something else entirely.
Last week, we brought together @AnthropicAI and @_StacksAI at our London office for “AI Agents in Production — Shipping AI that actually works,” a candid conversation on what it takes to bridge that gap.
Adriaan Engelbrecht, Applied Engineer at Anthropic, highlighted the importance of evaluating AI models tailored to specific use cases rather than a one-size-fits-all approach. Ivan Bovyrin, Founding Engineer at Stacks, stressed that human-in-the-loop patterns are key to gathering customer feedback and improving agent performance.
Both speakers emphasized that trust comes from deterministic, reliable systems.
It was inspiring to be in the room alongside the engineers and builders at the forefront of Applied AI. Conversations like these continue to move the needle.
Stacks retweeted
We've been running the PMM function at @_StacksAI as a team, but it needs someone to own it full-time. We're hiring our first Product Marketing Manager, positioning, launches, and competitive intel, all from scratch. Real ownership, no layers, high bar for craft. London.
DM me.
ALT Stacks is hiring our first product marketing manager.
Stacks retweeted
Harmonic just dropped its Q2 '26 Hot 25 list, featuring the early stage companies seeing the most investor interest on @Harmonic_AI's platform.
My biggest takeaways:
Almost zero competition between these products. Every quarter, this list has multiple clusters of startups all sort of doing the same thing. This quarter you can argue that all of these products operate in completely different spaces / industries / core ICP.
11 of 25 were on Q1's list. This is the highest I've ever seen carry over between quarters.
23 of the top 25 have raised in the past year. On the same thread as above, investors love concentrating into the winners.
Very strong technical teams. If you look at the founders backgrounds (link in next post), there's a strong bias towards very technical founding teams.
🤝 Paid partnership
Most variance commentary stops at "what happened."
The best explains why it matters, and what to do next.
Christian Wattig (@Wharton FP&A) and @albertmalikov are going live April 2nd for a 60-min masterclass on AI-driven variance analysis.
You'll walk away with:
→ The "What, Why, So What" framework
→ The ARCTIC method for finding the real story
→ How to generate variance commentary in seconds with AI
Sign up link in comments 👇
Today we're launching Flux AI Agent.
Flux investigates variances the way a senior analyst would. It pulls transaction-level evidence directly from your ERP, enriches it with full context, flags outliers, and writes explanations with citations back to source data. What used to take days happens in seconds.