TickerTain
TickerTain
NewsroomShortsPortfolioConvergence
NewsroomShortsPortfolioConvergence
←
▶ 49:55 · $MERGE · Merge CTO warns token maxing fails; real productivity comes from cycle-time reduction
episode briefing
Sourcery VC

SemiAnalysis, Altimeter, Nebius, Glean.. 12 Hot Takes From Biggest Names in AI

2026-07-19 · 11 company · 8 thematic
sentiment
8 bull1 bear2 neu
speakers
mark

CTO of AMD, leading chip design using thousands of AI sub-agents; sees agentic AI driving 1:1 CPU:GPU ratio and transforming end-to-end workflows.

casser

Casser leads Applied Intuition, a 1,000+ engineer company building software to put intelligence on a billion machines across automotive, defense, construction, mining and maritime.

arvin

Arvin Jain leads Glean, an enterprise AI platform that provides context graphs and model routing to reduce token costs and accelerate AI adoption in large organizations.

nquille

Nquille built Turbo Puffer, a vector search engine on object storage powering retrieval for Cursor, Notion, Anthropic and other AI-native companies at 10x lower cost.

max

Max founded Legora, a legal AI company that raised $600M in two weeks and pioneered consumption-based pricing to align token costs with customer value.

gil

Gil is CTO of Merge, an API integration platform; he warns that token maxing fails and integration security is the primary blocker to enterprise AI deployment.

quote
“Token maxing not working. You're getting zero results from using more tokens. It was cool to encourage your employees to use more AI, but I think what we're seeing now is we're not getting more output. The only thing that people are seeing…”
—gil
now playing · $MERGE
$APPLIED-INTUITIONbullish· high· poscasser
Applied Intuition targets intelligence on billion machines across autos, defense, industrial
Applied Intuition's platform approach — training models once and deploying across cars, trucks, ships, mining equipment and construction machinery — creates massive leverage on ex…
$NVDA···neutral· highdylan patel
Nvidia B200/B300 pricing rising before production as memory costs flow through
Next-gen Nvidia server pricing has already increased significantly versus last generation, and quoted prices for unreleased hardware have risen pre-production, signaling memory co…
$NBIS···bullish· high· posmark
Nebius pursues diversified 20-data-center portfolio to avoid single-project bottlenecks
Nebius's multi-threaded data center portfolio across 20 sites mitigates the local regulatory, power and community risks that stall single-project competitors, while building from…
$GLEANbullish· high· posarvin
Glean's context graph cuts token costs by picking right model and delivering data in one shot
Glean reduces enterprise AI token spend two ways: routing tasks to cheaper open/closed models that still work, and its context graph eliminates repeated retrieval cycles so models…
$ALTIMETERneutral· high· posapor
Altimeter urges CIOs to plan for multimodel climate not seasonal lab leadership
Enterprise buyers should architect for a permanent multimodel world — routing 90% of volume to open source, reserving frontier models for coding and novel use cases — rather than…
$TURBO-PUFFERbullish· high· posnquille
Turbo Puffer's object-storage search cuts cost 10x but still too expensive for new products
By layering caching on S3/GCS, Turbo Puffer delivers near-RAM vector search performance at object-storage economics, powering Cursor, Notion and Anthropic; further 10x cost drops…
$WONDERFUL-AIbullish· high· posbarack
Wonderful AI bets geographic expansion over vertical focus for enterprise AI land grab
Countries matter more than verticals for applied AI because labor displacement and go-to-market dynamics are nationally regulated; Wonderful executes an Uber-style horizontal play…
$LEGORAbullish· high· posmax
Legora pioneers consumption pricing in legal AI after $600M raise in two weeks
Legora aligns revenue with token usage via consumption pricing, maintaining positive gross margins while flat-rate competitors bleed on heavy users; expects the model to become st…
$MERGEbearish· medium· posgil
Merge CTO warns token maxing fails; real productivity comes from cycle-time reduction
Simply encouraging more token usage yields zero output gains; only embedding AI to shrink feedback loops and iteration cycles drives measurable productivity.
$NAVANbullish· high· posariel
Navan hits $10B bookings, 50% usage growth and profitability with anti-hallucination travel AI
Navan's proprietary anti-hallucination model enables reliable agentic travel booking — a domain where errors are unacceptable — driving 40% revenue growth, cash-flow positivity an…
$MDB···bullish· high· poscj
MongoDB CEO declares data layer the unsung hero of AI application quality
AI application performance is bounded by data layer quality; MongoDB's unified vector search, embeddings and operational data platform eliminates pipeline stitching and makes it t…