TickerTain
TickerTain
NewsroomShortsPortfolioConvergence
NewsroomShortsPortfolioConvergence
$ANTHROPIC·$MA····$INTC····$BLUE-ORIGIN·$SPCX····$CRWV····$CRM····$MSFT····$NVDA····$ORCL····$CURSOR·$AAPL····$OPENAI·$AMZN····$UBER····$GOOGL····$META····$TSLA····$DATABRICKS·$PERPLEXITY·$LYFT····$NBIS····$TSM····$LITE····$ANDURIL·
$ANTHROPIC·$MA····$INTC····$BLUE-ORIGIN·$SPCX····$CRWV····$CRM····$MSFT····$NVDA····$ORCL····$CURSOR·$AAPL····$OPENAI·$AMZN····$UBER····$GOOGL····$META····$TSLA····$DATABRICKS·$PERPLEXITY·$LYFT····$NBIS····$TSM····$LITE····$ANDURIL·
←

rodrigo liang

T2 · manager / operator

32-year semiconductor veteran. Founded SambaNova in 2017 to build reconfigurable dataflow accelerators for AI. Led company through six chip tape-outs (SN10-SN50) and $2.5B total fundraising, most recently $1B at $11B valuation led by General Atlantic.

6 calls·5 names·83% bull·last heard 2 months ago·Bloomberg Tech+1
track record

no scored calls yet — needs a stated position or a categorical verdict, with a matured window vs SPY

top calls

highest conviction · one per company
1sthigh conviction
$SAMBANOVASambaNova Systemsposition

SambaNova raises $1B at $11B valuation to scale premium inference chips

SambaNova's reconfigurable dataflow architecture delivers 10x power efficiency vs Nvidia GPUs (10kW vs 140kW per rack), enabling trillion-parameter models in a single air-cooled rack. The $1B round led by General Atlantic funds scaling of SN40/SN50 chips for premium inference — large models at full precision with ultra-low latency — targeting enterprises, sovereigns, and neoclouds needing differentiated margins.

Sourcery VC2026-07episode →
2ndhigh conviction
$NVDANvidia

Nvidia GPU racks become commodity infrastructure; inference providers need differentiated silicon for margins

Nvidia dominates training but inference economics favor specialized accelerators. SambaNova's SN40 runs trillion-parameter models at 10kW vs 130-140kW for Nvidia, air-cooled vs liquid. Inference providers running 70-80% workloads on Nvidia can route to SambaNova to free up racks for HPC/training, boosting revenue per rack and margins. Nvidia becomes undifferentiated commodity — same hardware as competitors — while premium inference commands pricing power.

Sourcery VC2026-07episode →
3rdmedium conviction
$VECTOR-CORE-COMPUTEVector Core Compute (VC2)

Vista Equity and Cambium launch Vector Core Compute Neocloud with SambaNova for ultra-low latency inference

Vista Equity and Cambium's new Neocloud VC2 deploys SambaNova racks in ultra-low latency metro data centers. Partnership model lets SambaNova focus on silicon/racks while VC2 raises capital for facilities — capital-efficient scaling for sovereign and enterprise premium inference.

Sourcery VC2026-07episode →

most discussed · click a bar to filter

  • $SAMBANOVA
  • $VECTOR-CORE-COMPUTE
  • $NVDA
  • $GENERAL-ATLANTIC
  • $ARMADA

recurring themes

  • AI Hardware & Chip Architecture1
  • Data Center Infrastructure1
  • AI Infrastructure1
  • AI Agents1
  • AI Economics & Business Models1
6 total
$SAMBANOVA
SambaNova
HIGHrodrigo liang·Bloomberg Tech·3 months ago·Jittery Tech Markets; OpenAI Gets BofA Credit Line | Bloomberg Tech 7/08/2026· position
SambaNova raises $1B at $11B valuation for inference-focused dataflow architecture
SambaNova's SRAM-based dataflow architecture complements Nvidia GPUs for high-speed inference, enabling trillion-parameter models with mature HBM supply, targeting enterprise and cloud customers like JPMorgan.
"SAMBANOVA IS A DATA FLOW ARCHITECTURE WHICH IS SRAM BASED... WE USED H.P.M., H.B.M. THAT WAS ALREADY IN MATURE PRODUCTION WHICH ALLOWS US TO GENERATE SIGNIFICANTLY MORE SUPPLY VER…"
36:42
$VECTOR-CORE-COMPUTE
Vector Core Compute (VC2)
MEDrodrigo liang·Sourcery VC·2 months ago·Inference 101: SambaNova CEO Rodrigo Liang
Vista Equity and Cambium launch Vector Core Compute Neocloud with SambaNova for ultra-low latency inference
Vista Equity and Cambium's new Neocloud VC2 deploys SambaNova racks in ultra-low latency metro data centers. Partnership model lets SambaNova focus on silicon/racks while VC2 raises capital for facilities — capital-efficient scaling for sovereign and enterprise premium inference.
"Last month we announced this great partnership with Vista Equity and Cambium on this new Neocloud Vector Core Compute VC2... they're deploying these ultra low latency data centers…"
38:36
$NVDA
···
Nvidia
HIGHrodrigo liang·Sourcery VC·2 months ago·Inference 101: SambaNova CEO Rodrigo Liang
Nvidia GPU racks become commodity infrastructure; inference providers need differentiated silicon for margins
Nvidia dominates training but inference economics favor specialized accelerators. SambaNova's SN40 runs trillion-parameter models at 10kW vs 130-140kW for Nvidia, air-cooled vs liquid. Inference providers running 70-80% workloads on Nvidia can route to SambaNova to free up racks for HPC/training, boosting revenue per rack and margins. Nvidia becomes undifferentiated commodity — same hardware as competitors — while premium inference commands pricing power.
"Instead of a 130 140 kilowatt rack of Nvidia GPU, we were outperforming them with a 10 kilowatt SM40 rack... People forget as much as Nvidia costs is commodity. Right. Because wha…"
0:25
$SAMBANOVA
SambaNova Systems
HIGHrodrigo liang·Sourcery VC·2 months ago·Inference 101: SambaNova CEO Rodrigo Liang· position
SambaNova raises $1B at $11B valuation to scale premium inference chips
SambaNova's reconfigurable dataflow architecture delivers 10x power efficiency vs Nvidia GPUs (10kW vs 140kW per rack), enabling trillion-parameter models in a single air-cooled rack. The $1B round led by General Atlantic funds scaling of SN40/SN50 chips for premium inference — large models at full precision with ultra-low latency — targeting enterprises, sovereigns, and neoclouds needing differentiated margins.
"We just did a first close of a billion dollar fund raise at an 11 billion valuation... We released SN40 a couple years ago. It became incredibly popular because instead of a 130 1…"
0:00
$GENERAL-ATLANTIC
General Atlantic
MEDrodrigo liang·Sourcery VC·2 months ago·Inference 101: SambaNova CEO Rodrigo Liang
General Atlantic leads SambaNova $1B round at $11B valuation
General Atlantic led the $1B first close alongside Seligman Ventures, T. Rowe Price, and Capital Group — significant American crossover investors signaling conviction in SambaNova's inference scaling thesis and path to profitability for neocloud customers.
"The round was led by General Atlantic with a number of incredible investors that came in — Seligman Ventures, Troll Prize, Capital Group. These are all significant American invest…"
1:32
$ARMADA
Armada
MEDrodrigo liang·Sourcery VC·2 months ago·Inference 101: SambaNova CEO Rodrigo Liang
Armada partners with SambaNova for modular edge data centers in oil, mining, defense
Armada deploys shipping-container data centers (10-20 racks) for remote/edge sites — oil rigs, mines, military. SambaNova's 10kW/rack air-cooled SN40 fits 10x more compute per container vs Nvidia's 140kW liquid-cooled racks, enabling AI in power/space-constrained environments. Partnership validates distributed inference model beyond hyperscale.
"Armada's a partner of ours. They've been a partner for a few years now... if you can actually take instead of having to put a 100 kilowatt Nvidia rack in there, you can put a 10 k…"
23:00
9
AI Hardware & Chip Architecturetailwind
Reconfigurable dataflow architecture beats GPUs for inference: 10x power efficiency, full precision, single-rack trillion-parameter models
SambaNova's SN40/SN50 chips use a software-defined dataflow architecture (vs fixed-function GPU cores) to map entire models onto silicon without quantization. This delivers 10kW/rack vs 140kW for Nvidia H100, air-cooled deployment in standard 19" racks, and full BF16/FP16 precision at higher token throughput. Enables premium inference tier: largest models, fastest latency, lowest opex.
9
Data Center Infrastructuretailwind
Inference shifts data center topology from gigawatt centralized to distributed metro/edge: 10kW racks enable deployment in existing colos, shipping containers, sovereign sites
Training needs gigawatt clusters with liquid cooling and dedicated power (nuclear). Inference is scale-out: add racks as users grow. SambaNova's 10kW air-cooled racks fit in any existing data center (Paris downtown, metro metros, shipping containers on oil rigs). Eliminates 9-18 month build cycles for liquid-cooled gigawatt sites. Enables sovereign/enterprise on-prem and neocloud metro latency (<10ms) for agentic workloads.
9
AI Infrastructuretailwind
SambaNova's inference dataflow architecture complements Nvidia with SRAM+HBM
SambaNova's SRAM-based dataflow chips deliver 10x faster decode for inference, enabling trillion-parameter models using mature HBM supply, targeting enterprise and cloud customers who need high-speed, low-power inference alongside Nvidia GPUs.
8
AI Agentstailwind
Agentic workflows demand sub-100ms latency: 20-agent chains at 2s each = 40s user wait. Metro-deployed inference becomes mandatory
Single-model chat tolerates 2s latency. Agent orchestration (security, banking, travel, reporting) chains 10-20 models sequentially. At 2s per hop, user waits 20-40s — unacceptable. Requires <100ms per hop, meaning inference must run in metro data centers near users (banks in Manhattan, hospitals in Paris). SambaNova's 10kW racks enable this distributed low-latency fabric; gigawatt training clusters in West Texas cannot. Creates structural demand for premium inference at the edge.
8
AI Economics & Business Modelstailwind
Inference providers measure revenue per rack: tokens/sec × price/token × utilization. SambaNova raises rack revenue 3-5x vs Nvidia at lower opex, enabling neocloud profitability
Neoclouds buy racks on capex, monetize via token APIs. Revenue per rack = tokens/sec × $/token × 720 hrs/mo. SambaNova's 10kW rack generates more tokens on larger models (higher $/token) at 1/14th power cost. Providers running 70-80% inference on Nvidia can offload to SambaNova, freeing Nvidia racks for HPC/training (higher $/hr). Net: higher blended margin without new capex. This is the KPI driving adoption.
8
Sovereign AItailwind
Nations and enterprises build private models on owned infrastructure to protect IP and data; SambaNova enables sovereign training+inference stacks
Japan, Korea, EU, and global banks/enterprises are training national/private models to avoid data leakage into frontier models (e.g., bank PII in ChatGPT, Figma designs in Anthropic). They need full-stack infrastructure they control — not rented from US hyperscalers. SambaNova sells racks/chips/software for on-prem/partner clouds, enabling sovereign fine-tuning and inference without data leaving jurisdiction. TAM expands beyond hyperscalers to 200+ sovereign/enterprise buyers.
7
Neoclouds & Cloud Computingtailwind
Neoclouds differentiate via silicon heterogeneity: 2-4 chip architectures (Nvidia + SambaNova + AMD + custom) replace single-vendor commodity stacks
Neoclouds A and B both offering Nvidia H100s are undifferentiated — same hardware, same pricing, margin compression. Winners will blend 2-4 architectures: Nvidia for training/HPC, SambaNova for premium inference, AMD for cost-sensitive, custom for sovereign. This heterogeneity lets them offer tiered services (ultra-low latency, sovereign, high-throughput) at different price points, raising blended margins. SambaNova's partnership model (not building competing cloud) accelerates neocloud adoption.