Partner
CoinFundCheck size: No verified personal check-size range is published. CoinFund attributes pre-seed, seed, and Series A deals to Einar, but disclosed round totals are company financings and not his personal allocation authority.
Einar backs foundational technology that decentralizes scarce AI resources or rebuilds financial rails onchain. His portfolio connects open compute and model ownership, verifiable AI, permanent data, cross-chain liquidity, stablecoin payments, and institutional prediction-market infrastructure. He favors technically ambitious early teams addressing a structural bottleneck, with a credible path from an initial developer or institutional wedge to an open network that scales access, freedom, and knowledge.
Outlier founders with exceptional technical depth, vision, and execution speed; early products at the foundational layer; open networks that broaden access to compute, intelligence, data, or finance; a sharp bottleneck and large market; credible developer or institutional demand; and teams willing to build difficult infrastructure before the opportunity becomes consensus.
No explicit personal pass list is public. His published theses imply weaker fit for incremental applications without a foundational advantage, decentralized systems with no reason to outperform centralized alternatives, AI projects without compute or data differentiation, financial infrastructure lacking institutional-grade liquidity and compliance, and technical networks without developer usability or real demand.
Use a warm introduction where possible, his stated preferred channel. Lead with the foundational bottleneck, why decentralization or onchain architecture creates a real advantage, and the team's technical edge. Bring benchmarks, developer or institutional usage, network design and incentives, security, unit economics, regulatory path, and the expansion loop. For AI, explain compute, data, model ownership, verification, and open-source distribution; for finance, quantify liquidity, settlement, compliance, and capital efficiency.
Identify the scarce centralized resource or control point, show how an open network improves access, ownership, verification, resilience, or economics, and prove that coordination costs do not erase the advantage.
Start with a specific developer or institutional bottleneck, validate technical performance and urgent demand, then map how network participation, integrations, and adjacent services turn the wedge into an infrastructure layer.
For new financial rails, test liquidity depth, cross-venue settlement, custody, credit, compliance, reliability, capital efficiency, and the ability to serve professional flows before assuming consumer-scale adoption.
Prediction markets need institutional-grade banking, settlement, credit, and cross-venue capital movement to deepen liquidity and mature into a durable asset class.
A programmable permanent-data layer can become decentralized storage infrastructure for AI and applications by combining verifiability, low cost, and data-native execution.
Decentralized training can advance AI research and preserve models as open public goods by coordinating compute and ownership outside a handful of centralized labs.
A native liquidity layer can turn Telegram's vast distribution into onchain economic activity when product value comes first and social and viral mechanics amplify it.
No podcast appearances are linked yet. Ask Kit to check for interviews that reveal how this investor thinks.
“The core promise of prediction markets is the ability to harness the wisdom of crowds, turning speculation into collective insight.”
— https://www.coinfund.io/insights/edge_markets_series_A
Einar is a Partner at CoinFund and a member of its executive team. Before CoinFund he invested at Accel and General Catalyst and worked at Mosaic Ventures and Bain & Company's private-equity group; CoinFund credits him with helping facilitate the $67B Dell–EMC transaction. Originally from Norway and based in the United States, he earned a BA from the University of St. Gallen and a Master in Management from London Business School, with study at Harvard.
Aggregating fragmented GPU supply and enabling collective model ownership can broaden access to frontier AI development and counter concentration in compute and foundation models.
Zero-knowledge machine learning can make AI inference verifiable and usable inside smart contracts, enabling autonomous onchain applications without trusting a centralized model operator.