AI is making building and launching products exponentially faster, but current user research is still painfully slow, expensive and biased. Even with AI-moderated interviews, getting user feedback from 50 – 100 participants still takes days, which is slow when you’re iterating on your product’s development and launch in hours. So companies skip user research at many important steps, which leads to AI slop.
Vectorial believes the next frontier in user research is every company having a living model of its target customers, modeled after thousands of customers, that keeps up to date with new emerging behavior and offers speed, breadth and depth in user research.
The company today announced a $3 million seed round backed by Industry leaders and institutional investors (Smilegate, SenseAI Ventures) and UC Berkeley ecosystem (Berkeley SkyDeck, Berkeley AI Research Lab), Stanford-led funds (Stanford GSB fund, Dent Capital) with participation from product leaders from Meta, Shopify and Mastercard. The round reflects the company’s deep academic backbone with AI ecosystems of Stanford, StartX, CMU (Venture Bridge) and UC Berkeley.
The founders of Vectorial are making a bold bet: within four years, running a behavior simulation with audience models before building or launching a consumer product or agent will be the norm. Simulations have been common in the aerospace, manufacturing and chip industries, yet the consumer products that touch billions of people still launch on gut instinct and a few experiments with live users.
Vectorial’s human behavior simulation platform for software products and digital experiences lets product, design and marketing teams watch their model of target customers react to a product journey, campaign design or AI agent before a single real user touches it. These modeled users are powered by SAPIENS, a behavior model that learns the latent traits (socio-demographic, psychographic and behavioral traits) of the target customer. It is 30 percent more accurate in predicting behavior than general-purpose LLMs, outperforms academic behavior models such as Be.FM and Centaur, and is validated against real-user studies. Vectorial has already modeled more than 2.5 million real people and run more than 30,000 simulations.
Existing leaps in user research fall short. AI-moderated interviews have become the norm, but every new question means a multi-day study with a handful of participants, which is slow, expensive and biased. At the other extreme, simulated user agents answer instantly but use synthetic users with no grounding in real behavior data, which doesn’t build trust with customers. Other simulation companies focus on market research to support strategic decisions, but they don’t help with downstream decisions or put users inside the product development and launch loop. By contrast, Vectorial’s audience is grounded in real behavior data at massive scale and fits inside the product development, launch and build workflow.
Vectorial’s founding team comes from Stanford, CMU, ISB and BAIR, with deep domain expertise in understanding and modeling behavior and how it applies to business today. Taran has spent a decade modeling human behavior, including in defense research and research labs. Rahul, who launched multiple products at unicorn companies, worked on a behavior graph for sales at his last startup, and Jasmine led synthetic user for marketing at Pfizer.
Vectorial’s customers span healthcare, gaming and e-commerce and include public companies and billion-dollar enterprises serving enterprises across the globe – like MIDI health, GoodRx, Suki AI , Boat, Everkind, Concrete software and many more .. Customers run around 600 simulations a month (40 users studies/ month), 20x more research than before, and simulations have reproduced prior user-study findings with more than 85% accuracy.
Vectorial will use the cash raised to expand its behavioral datasets through new data partnerships and advance SAPIENS 2.0 to become the de facto generative simulation infrastructure for building and launching products.
Abstract Expands Its Agentic SIEM: Security-Focused Models, Real-Time Triage, Response on Every Detection
Posted in Commentary with tags Abstract Security on October 20, 2026 by itnerdAbstract today announced new AI SOC capabilities in its agentic SIEM: investigation, hunt, and triage powered by a combination of Abstract’s own trained and tuned models and frontier foundation models, and SOAR response capabilities with actions assigned directly to detection rules. The capabilities are built into the platform that already collects, routes, and detects on security data in stream.
Security teams are facing agentic attacks that move at machine speed while their own tooling spreads the response across a pipeline vendor, a SIEM, a SOAR, a threat intelligence feed, and an AI triage bolt-on. Dozens of AI SOC vendors sell agents that sit on top of someone else’s alerts and add another contract to the stack. Abstract runs the entire loop on one platform: 400+ integrations feed streaming detection that operates in real time, before data lands, findings combine into entity-based insights, and the SIEM itself now investigates, classifies, and responds on the same data plane.
Most AI bolted onto the SOC re-enriches the same alert from scratch every time it looks at one. Abstract’s pipeline builds an enterprise context layer that makes the data AI ready before the alert ever fires: events are normalized and enriched in stream with identity, threat intelligence, and asset context. That normalization gives AI the security semantic layer it needs, so a field like src is never confused between the sender of the data, the host that opened the connection, or the internal host another source means by it, and every investigation starts from context the platform already trusts.
The agentic SIEM works every alert in the queue. Alerts that once aged out or got closed in bulk to keep the queue moving now each get an investigation with evidence attached, and analysts review conclusions instead of raw alerts.
The new capabilities span the full security operations loop:
Where foundation models trained on the open internet bring broad knowledge and unvalidated confidence, Abstract’s models ground each assessment in the ASTRO team’s validated threat expertise, so verdicts reflect how attacks actually unfold rather than how they read in public write-ups. Foundation models handle the deep reasoning on cases that warrant it, with the full evidence trail preserved.
Availability
The new AI SOC capabilities are generally available today as part of the agentic SIEM. Investigation bundles are available as add-ons sized by alert volume. Existing customers can add investigation bundles through their Abstract account team. Details at https://abstract.security.
Leave a comment »