Czar

Golf and #Bitcoin

[Charlie Zvibleman @CharlieZvible]

Excited to announce Alphasense’s latest funding round and crossing $600m ARR.

I’m also flat out giddy to officially announce SuperAnalyst, our agentic platform, along with the release.

The first artifact I created came up with an est for KR SSS & GMs by pulling historicals from Canalyst models, building 2 year / 3 year stacks, sending an agent to go back through the last few years of results and annotate impacts (calendar, weather, etc), pulled the AlphaSense channel checks, and triangulated to est for the upcoming quarter.

I then asked it to research key debates on HD and spin up subagents to research, loop, and weigh evidence on each of the top 3 debates. We’re able to automate the monitoring / push of new information coming out impacting those key debates due to our indexing of the data.

Our vertical integration, architectural choices, and focus on context engineering allow this to be accomplished with high token efficiency.

Efficiency is something you’ll hear a lot more. A 5-10x efficiency edge was nice in chat but compounds in long running tasks.

With the explosion of agents, efficiency literally becomes intelligence. A more efficient system can run more searches, test more hypotheses, call more tools, and verify more claims. In multi-step agentic workflows, strengths compound just as quickly as weaknesses. As accuracy and comprehensiveness build across each step, so do errors and blind spots. That dynamic gives SuperAnalyst an exponentially widening advantage, powered by the market's leading search foundation.

We’re using the capital to double down on investment in creating the most intelligent system. Building AI optimized tools and reinvesting our NNARR into our flywheel of expert calls to constantly close information gaps.

[Charlie Zvibleman @CharlieZvible]

Big product release from AlphaSense begins rolling out today: GenSearch v4. New architecture, better quality.

GenSearch / Deep Research begins to tool call our Financial Data (for the buyside, Financial Data itself also now rolls out). AS has a lot of disparate component parts (Canalyst models, screeners, read-throughs, channel checks) — this is the beginning of those parts being built into a cohesive AI analyst, where having more / better component parts, vertically integrated, will lead to better analysis.

Scheduled Agents: this materially upgrades monitoring. Put in the 10 theses you track and see a daily update on what came out that impacts each thesis. Soon, this will evolve from daily / weekly updates —> real-time AI monitoring.

For those of you with full page prompts, the system will move to break that into component parts (more compute) vs. trying to 1-shot an answer.

For your read-throughs: beta testing shows ~50% more tokens per query in v4 vs. v3. On top of that, another step change in usefulness creates a step change in usage.

[Czar @Czar213]

@Divergent7651 Try @AlphaSenseInc

[Charlie Zvibleman @CharlieZvible]

Cool -- AlphaSense is consuming as many tokens as all of Salesforce

[Charlie Zvibleman @CharlieZvible]

Solving the market debate

Will people pay for AI

Overlay of token consumption vs net new ARR growth at AlphaSense for your read-through… 😮

[Charlie Zvibleman @CharlieZvible]

I’m a little late but excited to announce our acquisition of Carousel and welcome Daniel, Jude, and team to AlphaSense.

We think the market for financial modeling in Excel evolves in a similar way to Cursor / IDEs – and is 2 years or so behind that market. You should be using an LLM to accelerate your modelling in Excel, and to do so, you need to learn what works / what doesn’t. 

AlphaSense, through Canalyst, knows the modeling workflow better than any tech company in the world. We sell a database of >4k fully drivable, hedge fund-quality models. That makes us uniquely positioned to break down that workflow into component parts and accelerate it with AI (Canalyst is training data for Carousel). For a simple example, building a retailer model based on a detailed sss + store build vs. SaaS model on rep productivity is a choice our LLM-based planner can make to deliver better models to end users. 

For Public companies, we will have a 100% accurate data repository that we can call -- which is faster and more accurate than any other method. This also sets up the most detailed and accurate evals as we automate and externalize financial data extraction to any private company. Automated updates of your models are coming. 

We will pull comps, multiples, stock prices, market data etc – as callable components from our (new) Financial Data/Excel plugin offering into Carousel. AI can learn to use our Excel plugin formulas. This will benefit heavily from vertical integration. 

Carousel is already winning over some impressive firms, but you can expect us to invest heavily into end user workflows in Excel and PPT. With enough thinking time (+ some scaffolding), we think some pretty interesting problems can be solved.

[Czar @Czar213]

LEAVE HIM IN!

[Charlie Zvibleman @CharlieZvible]

Latest AI read-through per request: usage continues to inflect. 

Frame of reference, AlphaSense token consumption is ~1% of MSFT+Azure - which an investor pointed out, puts this towards the tip of the spear of the end-user AI debate in the mkt.

AS Deep Research already offers the most valuable bundle of tokens per query I’m aware of. But to get ahead of the data chasers: as we extend thinking time - the next leg of token consumption is already clear.

[Czar @Czar213]

@NOTdaRealGil Elite. I caddied there for 9 years.

[Charlie Zvibleman @CharlieZvible]

This is the point where AlphaSense hits escape velocity.

After a massive effort iterating, a few weeks ago we quietly launched fully AI-hosted Expert Calls. Feedback has been incredible.

AlphaSense created a self-reinforcing flywheel where context from our system feeds a smarter AI-host. You’ll be amazed at the industry context, follow up questions etc - that come from having the context of all the prior calls synthesized by our Deep Research.

A few years ago AlphaSense acquired an early but exciting business called Stream. They had a built in flywheel of buyside analysts hosting calls with experts – in exchange for charging only cost for the call, the call was transcribed, reviewed by a professional compliance team, and published in a library. This content set truly grew like wildfire – with ARR up >20x since acquisition.

Then, a year ago we acquired Tegus – the leader / gold standard in the space. >50% of the midas list hosts their calls on Tegus. I am actually continually shocked at the caliber of investors hosting calls.

All of a sudden AlphaSense + Tegus grew to a library >200k, on pace to hit 9k / month shortly.

Then came AI – LLMs have flipped from dumb to smart and can now easily make sense of the library. That then becomes a smarter set of tokens to feed in to the system → which then leads to better calls (by both clients and AI) → which leads to better information in the platform —> which leads to better context …

Have no fear – the AI led calls will be separately labeled and investor led will continue to grow rapidly (just like Twitter – some people love the game [100x more love consuming]). But our search system essentially has the mind of the market and can now automatically fill the gaps.

We’re in Alpha testing on externalizing the system to let clients host their own AI-led Expert Calls – reach out if you’d like to test.

[Czar @Czar213]

@buccocapital I tested your version yesterday out in AlphaSense, great prompt. I’ll check this one out.

[Czar @Czar213]

@buccocapital Try this on @AlphaSenseInc

[Charlie Zvibleman @CharlieZvible]

Hyperscaler read-throughs -- how'd we do?

via AlphaSense Deep Research

[Charlie Zvibleman @CharlieZvible]

read-thru for $nvda earnings

[Czar @Czar213]

@baystate_golf @acaseofthegolf1 You’re 100% right Sean

[Czar @Czar213]

@CharlieZvible @AskPerplexity what is compute cost for each of these tweet responses?

[Czar @Czar213]

@baystate_golf Stratas from Young Lim

[Czar @Czar213]

@RampCapitalLLC 👀

[Czar @Czar213]

@ByeongHunAn @ByeongHunAn

[Czar @Czar213]

@NamethatYank Matt Nokes