Participated · Seed · Jul 2026

Cascade leverages AI to surface engineering, architecture and construction project opportunities by reading signals such as bond filings, permits, budget announcements, capital plans, property transactions, earnings transcripts and meeting minutes. The company says its platform has surfaced more than $10 billion in project opportunities for clients, including firms that build for major projects at JFK and LaGuardia airports. Cascade was founded by Hannia Zia and Joana Ferreira, who previously worked at Google and met at UnlikelyAI. The startup plans to use the seed funding to deepen its network and accelerate customer adoption. The company positions itself as reducing reliance on manual prospecting processes for architecture, engineering and construction firms. No revenue, valuation or prior financing details were disclosed in the article.

Participated · Seed · Jun 2026

Idilio produces vertical micronovelas with episodes of one to three minutes and seasons that can reach up to 80 chapters, aiming for strong hooks in the first three seconds. The company describes itself as data-first: it writes and produces initial chapters, publishes them in-app and on social, and uses rapid metrics to decide whether to rewrite or scale a story. Idilio reports roughly two million downloads in 120 countries and about 45 minutes of daily use per user, which the company says is double Netflix's mobile time by its metrics. Founder Gabriela Tafur, a Colombian lawyer, TV presenter and Stanford MBA, built the product after observing short-form dubbed dramas in mobile use. The company targets Latin American melodrama as a cultural differentiator versus Chinese-origin micro-series and aims to scale its audience internationally. The model emphasizes low-cost, fast production cycles but faces sector risks such as low retention and variable content quality.

Participated · Seed · Feb 2026

Ando is developing what it calls the world’s first AI infrastructure for the global hourly workforce, tackling the persistent mismatch between staffing levels and real-time demand. Its cloud platform ingests location-level data to forecast demand and auto-generate shift schedules that factor in each worker’s availability, preferences, and constraints. In pilots with enterprise customers over the past year, Ando’s models have achieved more than 90% daily demand-forecast accuracy and driven 100% conversion from presentation to pilot to paid contracts with full customer retention. Some client locations have reported $60,000–$80,000 in incremental revenue attributed in part to better staffing alignment. The company’s roadmap includes pushing forecasting granularity to 15-minute intervals to further optimize labor allocation. Fresh seed capital will fund product expansion and broader enterprise deployments. Prior to the seed raise, Ando had been operating with undisclosed pre-seed or founder capital; no other institutional rounds are mentioned.

Participated · Seed · Oct 2025

Founded in 2020 by Patrick Murphy, Stéphane Turbide, and Simon Vallée, Maket positions itself as the “ChatGPT for architecture,” embedding expert design logic into an intuitive generative-AI workflow. Users input basic project parameters and the software produces customised floor plans, allows real-time layout edits, and renders 3D visualisations, while offering AI-driven design recommendations. Key features include rapid floor-plan generation, style exploration, and an impending regulatory assistant that checks zoning code compliance; Maket 2.0, slated for early 2026, will add HVAC planning and material take-offs. The product is sold via subscription starting around $30 CAD per month, with larger enterprise tiers for builders and prefab manufacturers. Since launching publicly in 2023, the company has attracted more than one million registered users and grown to roughly 14–15 employees. To date it has raised over $4 million CAD in total funding, fuelled by a recent seed round.

Participated · Seed · Sep 2025

Manas AI builds neuro-symbolic, science-based foundational models that analyze biological data to discover and develop transformative medicines. Co-founded by Dr. Siddhartha Mukherjee, Reid Hoffman, and Ujjwal Singh, the company positions itself as an AI-native drug developer rather than a traditional biotech. Its core product suite consists of proprietary large-scale models designed to accelerate target identification, molecule generation, and pre-clinical validation. The newly raised capital will be directed toward further refining these models and advancing ongoing therapeutic campaigns. Since launch, the startup has secured a total of $50.6 million across two seed financings ($24.6 million in January 2025 and a $26 million extension in September 2025). Investors span top technology and life-science funds, signaling cross-sector confidence in Manas AI’s hybrid AI-biotech approach.