Czar retweeted
Today @AlphaSenseInc, we're launching the SuperAnalyst MCP, putting SuperAnalyst to work wherever your agents run. Rather than one more data pipe for your agents to brute force there way across, SuperAnalyst MCP allows them to delegate research to the world's best research system - thousands of sources, market-leading search, purpose built AI - so your agent spends its tokens on the problem you actually want solved.
We measure dramatically higher quality at 3x lower cost with ever claim cited to its source down to the sentence.
SuperAnalyst has become the platform for technology teams to automate complex investment research and build institutional grade market intelligence applications. Highly encourage folks to check it out: alpha-sense.com/resources/pr…
Czar retweeted
We’ve officially launched Channel Check Intelligence
The best new source of Idea Gen
AlphaSense began structured channel checks a bit over a year ago, already scaled to 3k per month (compliance reviewed, no embargo). Already much more than anyone could read but high signal if you can make sense of it.
Channel Check Intelligence makes sense of it —> what sectors, companies are inflecting, pricing / invy better or worse etc. Not only the numbers, but the why.
CCI is another form of pre-compute where we use continually use LLMs to structure the data once, for everyone. Soon, when you build your Earnings Previews with SuperAnalyst, the system will know to bring in CCI - and even use as an input to the assumptions in your model - without needing to rebuild the analysis from scratch (accuracy, time, tokens)
We are investing aggressively in scaling and development -- it's going to get pretty crazy in a year.
Czar retweeted
Great blog from AlphaSense AI team (@cackerso @spacemanidol ). Takeaways:
• OAI’s new models took a big step forward – GPT 5.6 Sol materially outperformed on answer quality while costing ~half (!) as much as Opus 5. Imagine this has something to do with their recent rumored surge.
• Open source models performed worse than GPT and Opus but mostly cost more. (headline prices don’t account for cost to serve many more tokens)
• Left alone, frontier models are bad searchers, just like a brilliant human navigating a library without a catalog. So they compensate with volume. Fetch, skim, re-fetch. Cost balloons, context dilutes, quality drops.
• Earlier this year, it became clear in our Evals there was no longer one “best” model. Models are better at certain things.
• Isolating this component of our stack, AlphaSense answers preferred 2.8:1 while being 3x cheaper – that is crazy valuable as long-running agents compound (search, optimized harness, model routing). This ignores using pre-built components.
• Frontier models aren’t going anywhere – but there’s now a pretty clear path to RL + Model Routing making another leap in performance (quality, cost) for specific tasks (search)
At this point, we’re selling a pretty huge bundle of compute and part of AlphaSense mission is to continue to optimize token efficiency → higher efficiency ultimately leads to higher accuracy (but the path there isn’t just cheap models). You’ll see us take some pretty interesting paths there enabled by our vertical integration – such as precompute.
Czar retweeted
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.
Czar retweeted
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 retweeted
Solving the market debate
Will people pay for AI
Overlay of token consumption vs net new ARR growth at AlphaSense for your read-through… 😮
Czar retweeted
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 retweeted
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 retweeted
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 retweeted
read-thru for $nvda earnings
Czar retweeted
Watching the last couple of Full Swing Episodes. Pardon me if this take has been thrown out in the buffet of takes…
Should Keegan Bradley be a Ryder Cup captain in the future? @NoLayingUp
Czar retweeted
VICTORY!!!!!
#GoCrossGo
Czar retweeted
Shannon Johnson defeats Julia Potter-Bobb in a battle of past champions.
Keep up with the rest of the action #USWomensMidAm ⬇️
Czar retweeted
Read a great Reddit thread putting historical dates in perspective.
Here are 8 gems.
1/ The moon landing was only 66 years after the Wright Brothers first flight (1903-1969). Within a lifetime, humans went from having limited flight tech to travelling ~239k miles from Earth.
Czar retweeted
Aaron Judge has hit four home runs in winner-take-all postseason games.
That's the most in MLB history.
Did @PReedGolf blank out the T on his White Hot putter?? @NoLayingUp @deep_fried_egg @bostonhomer @acaseofthegolf1
“Got a lot of golf left to play.. but I feel my game is where it needs to be!”
Round 2 tomorrow Bedminster! Looking forward to seeing you there! #LIVGolf
@LIVGolfInv
3
Chills
An inspiring story in Massachusetts. MassAm finalist Conner Willett lost his father over the weekend. He’s now playing in the 36 hole final on Friday. @shanebacon
If you’ve ever hit a putt on the 13th at Concord CC, you know how incredible this shot is to win his semi-final.
1
3























