Startup Beaker

Sapiom raised $15M seed to build payment infrastructure for AI agents.

Ex-Shopify payments director Ilan Zerbib spotted the gap: AI agents can't autonomously buy services like Twilio SMS or AWS compute. Every API call requires authentication and micro-payments that break agent workflows.

• $15M seed led by Accel
• Backers include Okta Ventures, Anthropic, Coinbase Ventures
• Targets B2B before consumer AI agent payments
• Handles authentication + billing for vibe-coding platforms

Founder steps:
• Leveraged 5 years at Shopify payments engineering
• Identified agent payment bottleneck before market matured
• Built enterprise-first approach (not consumer)
• Secured strategic investors across AI/fintech ecosystem

Takeaway: AI agent infrastructure needs financial rails, not just better models.

https://techcrunch.com/2026/02/05/sapiom-raises-15m-to-help-ai-agents-buy-their-own-tech-tools/

Brewed by Startup Beaker.

#fintech #AI

Fundamental raised $255M Series A at $1.2B valuation for structured data AI.

While LLMs handle text well, they fail on enterprise tables with billions of rows. Fundamental built Nexus — a Large Tabular Model that processes massive datasets without transformer limits.

• $255M led by Oak HC/FT, Valor Equity, Battery Ventures
• 7-figure Fortune 100 contracts already signed
• 20x faster inference vs traditional GPU clusters
• Strategic AWS partnership for deployment

Founder steps:
• Built non-transformer architecture for structured data
• Secured Fortune 100 pilots before major funding
• Partnered with AWS for distribution channel
• Demonstrated 20x performance benchmarks

Takeaway: Enterprise AI beyond chat is a billion-dollar opportunity.

https://techcrunch.com/2026/02/05/fundamental-raises-255-million-series-a-with-a-new-take-on-big-data-analysis/

Brewed by Startup Beaker.

#startups #AI

Fundamental just raised $255M Series A at $1.2B valuation for their "Large Tabular Model"

While LLMs dominate text, Fundamental built the first foundation model for structured data — the billions of tables that power enterprise predictions.

• $255M Series A led by Oak HC/FT, Valor Equity Partners
• $1.2B valuation emerging from stealth
• Already secured 7-figure Fortune 100 contracts
• Strategic AWS partnership for deployment

How it happened:
• Identified LLM gap → structured data unsolved at scale
• Built non-transformer architecture → handles billions of rows
• Pre-trained on enterprise tables → pattern recognition
• Secured Fortune 100 pilots → proved contract potential
• Landed AWS partnership → enterprise distribution

Takeaway: While everyone chases the next LLM, the biggest enterprise AI opportunity might be in structured data.

Source: TechCrunch

Brewed by Startup Beaker.

Anthropic just nuked $50B+ in legal tech market cap with one plugin launch.

Anthropic released Claude Cowork's legal plugin Feb 2nd — an agentic AI tool automating contract review, NDA triage, and compliance workflows.

Key features:
• Contract review with playbook methodology
• Document risk flagging and compliance tracking
• Redline generation with commercial rationale
• Enterprise system integrations

How they built it:
• Started with Claude LLM foundation
• Developed agentic Cowork for enterprise
• Created domain-specific legal plugins
• Built contract review workflows
• Released open-source for adoption

Market carnage:
• Thomson Reuters: -18% ($3.2B loss)
• RELX: -14% ($8.1B loss)
• Experian: -8.3% ($2.1B loss)

Takeaway: Foundation models are targeting specialized enterprise tools. Legal tech incumbents charging premium for basic AI just became vulnerable.

https://www.artificiallawyer.com/2026/02/02/anthropic-moves-into-legal-tech/

Brewed by Startup Beaker.

Ex-HubSpot execs just raised $20M from Sequoia to kill CRM as we know it.

Day AI calls itself "the Cursor of CRM" — an AI-native platform that learns your business in 15 minutes and acts as your chief of staff.

• $20M Series A led by Sequoia (Pat Grady joining board)
• Previous $4M seed also from Sequoia
• 120 customers after 1+ year private testing
• Founded by Christopher O'Donnell (ex-HubSpot CPO) & Michael Pici

How they built it:
• Identified self-updating CRM gap at HubSpot but lacked tech
• ChatGPT API launch in 2023 made vision possible → founded Day AI
• Claude 3.5 Sonnet breakthrough in June 2024 enabled full platform
• 15-minute onboarding vs traditional CRM complexity

Customers report saving accounts from churn and eliminating status meetings entirely.

Takeaway: When incumbents can't innovate fast enough, former insiders with fresh tech stacks can rebuild entire categories.

Source: Upstarts Media

Humans& just raised the 2nd largest seed round ever: $480M at $4.48B valuation.

AI lab founded by ex-Google, Anthropic, xAI, OpenAI, and Meta researchers building "human-centric AI" collaboration tools.

Key metrics:
• $480M seed round (all cash, unstructured)
• $4.48B valuation
• Founded September 2025 (4 months to unicorn)
• Led by SV Angel + Georges Harik
• Investors: Nvidia, Bezos, GV, Emerson Collective

How it happened:
• Assembled all-star team from top AI labs
• Focused on human-AI collaboration angle (differentiation)
• Leveraged founder networks for warm intros
• Raised unstructured to move fast
• Will spend majority on compute for model training

Takeaway: Star teams with clear differentiation can raise massive rounds in months, not years.

Source: Crunchbase News

Brewed by Startup Beaker.

4-month-old AI lab raises $480M seed at $4.48B valuation

Humans& just closed the second-largest seed round in history. Founded by AI researchers from OpenAI, Anthropic, xAI, Google DeepMind, and Meta.

• $480M raised in seed funding
• $4.48B valuation (unicorn status)
• Led by SV Angel + Georges Harik
• Investors: Nvidia, Jeff Bezos, Google Ventures
• Focus: "human-centric AI" for collaboration

How it happened:
• Sept 2025: Assembled top AI talent from 5 major labs
• Oct 2025: Defined contrarian thesis (AI for connection, not replacement)
• Dec 2025: Secured leads through co-founder's network
• Jan 2026: Closed oversubscribed round with tier-1 investors

The contrarian bet: while others build AI to replace humans, Humans& builds AI to connect them.

Brewed by Startup Beaker.

Ideavize.ai just cracked the innovation roadmap problem.

This generative AI platform launched on Product Hunt (Jan 22) and transforms how startups move from idea to execution. Built by Huebits Tech, it generates complete project roadmaps across 50+ domains.

• Clear problem statements with context
• Tailored technology stack suggestions
• Step-by-step implementation roadmaps
• Market intelligence with growth projections
• 1B+ potential innovation pathways

How it happened:
• Founded Huebits Tech in 2024 (India)
• Built domain-specific AI vs generic assistants
• Integrated technical + market intelligence
• Launched on Product Hunt for validation
• Opened SF office for US expansion

Why it matters: Most founders waste months on fragmented planning. Ideavize.ai delivers structured roadmaps that compress ideation-to-execution cycles.

Source: Huebits Tech

Brewed by Startup Beaker.

CVector raised $5M to build an AI "nervous system" for heavy industry.

The NYC startup creates software that translates micro-actions (like valve adjustments) into measurable savings for manufacturers.

• $5M seed led by Powerhouse Ventures
• Serves utilities, chemical producers, advanced manufacturing
• Team of 12, Manhattan office
• Customers include metals processing, ammonia production

How founders Richard Zhang and Tyler Ruggles built it:
• Started with utilities as wedge market
• Positioned as "operational economics" - bridging operations with profit
• Proved ROI with concrete examples
• Hired data talent from finance/hedge funds
• Capitalized on AI adoption shift in industrial sector

Takeaway: Industrial AI wins when it translates actions into CFO-measurable savings.

Source: TechCrunch

Brewed by Startup Beaker.

Synthesia just hit $4B valuation — nearly doubling in 12 months.

The London-based AI video platform raised $200M Series E, jumping from $2.1B to $4B valuation. Unlike most AI startups burning cash, Synthesia crossed $100M ARR in April 2025.

• $200M Series E led by GV (Google Ventures)
• $4B valuation (up 90% from $2.1B in 2025)
• $100M+ ARR with enterprise clients like Bosch, Merck, SAP
• Structured employee secondary sale via Nasdaq

How it happened — Founder steps:
• Started with AI avatar tech for corporate training (2017)
• Focused on enterprise wedge vs consumer market
• Built sticky B2B product with measurable ROI
• Crossed $100M ARR milestone (April 2025)
• Existing investors doubled down in Series E

Takeaway: Enterprise AI with clear ROI metrics attracts repeat investment at premium valuations.

https://techcrunch.com/2026/01/26/synthesia-hits-4b-valuation-lets-employees-cash-in/

#startups #funding

Brewed by Startup Beaker.

Two devs turned basement crypto miners into $120M ARR AI cloud platform.

Runpod started when Zhen Lu and Pardeep Singh's Ethereum mining became "boring" in 2021. Instead of selling their $50K GPU setup, they pivoted to AI hosting.

Key metrics:
• $120M ARR (4 years post-launch)
• 500K developers
• $20M seed (Dell + Intel Capital)
• Bootstrapped to $24M before funding

The playbook:
• Repurposed mining rigs → AI servers
• Posted on Reddit: "Free GPU time for feedback"
• Beta users → paying customers in 9 months
• Hit $1M revenue, quit day jobs
• Revenue-share partnerships (no debt)
• VC found them on Reddit

Takeaway: Sometimes the best distribution is the simplest.

Source: TechCrunch

Brewed by Startup Beaker.

Yuki just emerged from stealth with $6M seed led by Hyperwise Ventures.

What it is: AI-powered data cost optimization platform for Snowflake, BigQuery, and Iceberg data lakes.

The numbers:
• $6M seed round
• 42.6% average cost savings for customers
• 15 employees across Israel, US, UK
• Founded in 2025

How they built it:
• Founders (childhood friends) reunited after decade apart
• Both working in data space independently
• Identified gap: no control plane for data management
• Built metadata-only solution with enterprise security
• Revenue model: percentage of actual savings generated
• Landed customers like Tenable and Angel Studios

The insight: "Data is the only resource no one truly manages" - CEO Ido Arieli Noga

Brewed by Startup Beaker.

Skild AI just raised $1.4B to scale robotics foundation models.

SoftBank led the round. Valuation: $14B+. One of the largest robotics funding rounds ever.

What it is:
• Omni-bodied AI that controls any robot type
• One "brain" works across humanoids, quadrupeds, robotic arms
• No retraining needed for different robot bodies
• Already deployed in security, warehouses, manufacturing

How it happened:
• 2023: Founders left Carnegie Mellon with robotics research
• Built foundation model using human video data + physics simulations
• Secured early enterprise pilots (security patrols, warehouse inspection)
• Hit $30M revenue within months in 2025
• Attracted strategic investors: Samsung, LG, Schneider, Salesforce

Why this matters: Foundation models transformed language AI. Skild AI is doing the same for robotics. Instead of custom programming each robot, one AI adapts to any physical form.

The takeaway: Generalization beats specialization in AI. Build systems that work across use cases, not just one narrow application.

Source: VentureBurn

Brewed by Startup Beaker.

How to Fundraise in 2026: A Forum for Investors and Startups in Tashkent

On April 3, 2026, the 5th anniversary Central Eurasian Venture Forum will take place in Tashkent - a platform where investors, venture funds, banks, corporations, and tech startups from Central Eurasia and global markets meet.

🎯 Forum agenda: The Next Frontier for Smart Capital
A conversation about how investment, venture capital, and the AI market are evolving and where, in this new reality, strong partnerships and growth are being built.

✅ Who it’s for:
— investors looking for deal access
— banks, corporations, fintech companies
— startups aiming to raise capital in 2026

🔥 Early Bird −30% is available on all ticket categories until the end of January
Lowest price!

🎫 Get your ticket
📍 Tashkent | April 3, 2026

Telegram | Instagram | Linkedin | Facebook | YouTube

INVESTMENT READINESS ACCELERATOR — QABUL BOSHLANDI

Tashkilot: MOST C-Space Hub tomonidan, SQB Ventures (bosh hamkor) va UNDP ko‘magida tashkil etilmoqda.

O‘zbekistondagi pre-seed va seed bosqichidagi startaplar uchun 12 haftalik gibrid akseleratsiya dasturi. Bu g‘oya bosqichidagi loyihalar uchun emas, balki masshtablashga tayyor va institutsional kapital jalb qilishni maqsad qilgan jamoalar uchun mo‘ljallangan amaliy tizimdir

Siz Nima olasiz

Kapital: Eng yaxshi natija ko‘rsatgan ishtirokchilar uchun 200 000 dollargacha investitsiya jalb qilish imkoniyati.
Bitim strukturasi: Investitsiya raundlariga real tayyorgarlik, SAFE shartnomalari va yuridik hujjatlashtirishni o‘rganish.
O‘sish mexanikasi: Asosiy e’tibor nazariyaga emas, balki MRR (oylik daromad) va biznes metrikalarini oshirishga qaratiladi.
Networking: Investorlar bilan to‘g‘ridan-to‘g‘ri aloqa va venchur fondlar ishtirokidagi Demo Day.
Natija: Ushbu metodologiya bitiruvchilarga avval 700 mingdan 4 million dollargacha (masalan, Cerebra AI, Garderob) mablag‘ jalb qilishga yordam bergan.
Tashkilotchilar
Dastur MOST C-Space Hub tomonidan, SQB Ventures (bosh hamkor) va UNDP ko‘magida tashkil etilmoqda. Mentorlar qatorida MOST Holding va Justart VC Studio rahbarlari bor.

Talablar
Bosqich: Pre-seed yoki Seed.
Traction: Kamida 1000 AQSh dollari miqdorida oylik daromad (MRR) va ishlayotgan MVP.
Jamoa: Kamida 2 kishi.
Qatnashish
Arizalar 9-fevralga qadar qabul qilinadi. Quyidagi havola orqali ro‘yxatdan o‘ting:
https://investready.uz/uz

Humans& just raised the 2nd largest seed round in history.

The 3-month-old AI startup founded by Anthropic, xAI, and Google alums secured $480M at a $4.48B valuation.

Key metrics:
• $480M seed round (2nd largest ever)
• $4.48B valuation in 3 months
• 20+ employees from top AI labs
• Investors: Nvidia, Jeff Bezos, GV

How they did it:
• Assembled dream team from Anthropic, xAI, Google
• Positioned as "human-centric AI" vs replacement narrative
• Leveraged founder pedigree for pre-product funding
• Built AI collaboration platform with memory

Takeaway: Team pedigree + contrarian positioning unlocks massive pre-product funding.

Source: TechCrunch

#startups #funding

AI lab Humans& just closed a $480M seed at $4.48B valuation in 3 months.

What it is: Human-centric AI for collaboration, founded by ex-OpenAI/xAI/Google/Anthropic researchers.

Key metrics:
• $480M seed round
• $4.48B valuation
• 3 months from founding to close
• Led by SV Angel + Georges Harik
• Nvidia, Bezos, GV participated

How it happened:
• Assembled all-star team from top AI labs first
• Georges Harik (Google employee #7) as co-founder/lead investor
• Eric Zelikman brought xAI/Grok experience
• Andi Peng added Anthropic Claude expertise
• Positioned as "human empowerment" vs replacement narrative
• Leveraged team pedigree for massive pre-product valuation

Takeaway: In AI, team pedigree can unlock billion-dollar valuations before product launch.

Brewed by Startup Beaker.

Rain just raised $250M Series C at $1.95B valuation — 17x growth in 10 months.

What it is: Enterprise stablecoin payments infrastructure that makes crypto feel like traditional payments.

Key metrics:
• $250M Series C led by ICONIQ
• $1.95B valuation (17x in 10 months)
• $3B+ annualized transaction volume
• 200+ enterprise partners including Western Union
• 30x active card growth in 1 year

How it happened:
• Built full-stack infrastructure (cards, wallets, compliance)
• Became Visa Principal Member for global acceptance
• Focused on enterprise-grade compliance from day one
• Partnered with major players like Western Union early
• Scaled to 150+ countries with regulatory readiness

The takeaway: Stablecoins are becoming mainstream payment rails. Rain positioned as the infrastructure layer enterprises need.

Source: PR Newswire

Brewed by Startup Beaker.

Skild AI just closed a $1.4B Series B at $14B valuation.

What it is: Pittsburgh-based "robot brain" startup building omni-bodied AI that controls any robot without prior training.

• $1.4B raised, SoftBank led
• $30M revenue in 2025 (profitable)
• 14B+ valuation in 3 years
• Bezos, NVIDIA, Sequoia backed

How it happened:
• Started 2023 with physics simulation training
• Built "omni-bodied" learning (adapts to any robot form)
• Proved concept with humanoid/dog/arm robots
• Landed enterprise pilots (security, warehouses, data centers)
• Scaled to $30M ARR before mega-round
• Expanded globally (SF, India offices)

Takeaway: Train on simulations first, prove with real deployments, then scale fast.

Source: Technical.ly

Brewed by Startup Beaker.