Vom «Rundschau»-Moderator zum Podcast-Host: Brotz plant Neustart bei Blick
Sandro Brotz kehrt dem Schweizer Radio und Fernsehen den Rücken und wechselt Anfang 2027 zur Boulevardzeitung «Blick». Der 56-jährige Journalist, derzeit Moderator und stellvertretender Leiter...
Roche beschleunigt Entwicklungszeiten und rückt wachstumsstarke Projekte in den Vordergrund
Roche versucht nach einer Phase von Forschungsrückschlägen und Belastungen durch auslaufende Patente, das Tempo in der Medikamentenentwicklung deutlich zu erhöhen. Am Pharma-Investorentag...
Höhere Energiepreise treiben Inflation – Anleger ziehen sich aus Aktien zurück
Inflationssorgen und steigende Ölpreise haben Europas wichtigste Aktienmärkte zur Wochenmitte spürbar belastet und zugleich die Monatsbilanz eingetrübt. Der EuroStoxx 50 schloss am Mittwoch...
Parmelin eröffnet Olma mit Ferkel – und warnt vor schwieriger Lage der Bauern
Mit einem betont unterhaltsamen und zugleich programmatischen Auftritt hat der Kanton Schwyz seine Rolle als Gastkanton an der 83. Olma in St. Gallen übernommen. Bereits die offizielle...
Mutmassliche Brandstiftung in Baar: Sachschaden von mehreren hunderttausend Franken
Nach dem nächtlichen Grossbrand an der Ruessenstrasse in Baar im Kanton Zug gehen die Strafuntersuchungsbehörden von vorsätzlicher Brandlegung aus. Mehrere Fahrzeuge waren am späten Abend des...

Toptal Launches ExpertCrowd to Train AI with Vetted Domain Experts, Releases Frontier SWE Benchmark

13.10.2026

SAN FRANCISCO, Oct. 13, 2026 /PRNewswire/ -- Toptal, the world's largest talent network of highly skilled domain experts, today announced the launch of ExpertCrowd, a platform for training and improving AI models and systems through RLHF, RL data generation, data labeling, and model evaluation. ExpertCrowd combines its advanced platform for AI model development and testing with Toptal's global network of specialized talent to help frontier AI labs, innovative enterprises, and fast-growing startups build more accurate and capable AI models and systems.

The platform enables organizations to move beyond generalized crowdsourcing and data annotation and apply highly specialized human judgment to develop and improve AI applications and models across industries including financial services, healthcare, legal, biotechnology, cybersecurity, and more. ExpertCrowd is powered by Toptal's global talent network, with more than 30,000 specialized domain experts across 140+ countries, including more than 2,000 AI and big data specialists, 12,000 professionals with advanced degrees, and 1,000 PhD-level AI specialists, available on demand.

ExpertCrowd matches thousands of specialized experts to AI alignment and training projects instantaneously, enabling organizations to rapidly scale such projects with confidence. Its AI-driven sourcing engine was developed in-house by Toptal's AI research team, composed of more than a dozen PhDs and researchers from leading institutions such as Harvard, EPFL, and the University of Liège, using 15 years of Toptal's proprietary data in sourcing, screening, and matching highly skilled domain experts.

Offering multimodal annotation with customizable workflows and live project tracking, ExpertCrowd provides real-time progress updates as domain experts do the actual work. AI researchers applying GRPO, PPO, and other RL techniques have leveraged ExpertCrowd's platform to further the development of their AI models and systems.

Prior to today's official launch, ExpertCrowd generated over $5 million in recurring revenue from work with frontier and neo labs, as well as companies including Scale AI, Sterling Labs, Vetto, and other innovative startups that service frontier labs.

As part of the launch, ExpertCrowd has developed and published the ExpertCrowd-SWE benchmark, which evaluates SOTA AI coding agents on previously unseen issues. The ExpertCrowd-SWE benchmark exposes the most significant deficiencies in model capability relative to any other currently published SWE benchmark. ExpertCrowd-SWE spans 87 codebases in 11 programming languages. Engineers from Toptal's talent network led the development of this benchmark and authored every task as well as each grading rubric. Researchers, including several PhDs in various areas of artificial intelligence from Toptal's AI team, reviewed and assessed the benchmark's methodology and results.

"The fact that our SWE benchmark results expose the most deficiencies relative to any other published benchmark on leading models such as Opus 5.5 and Fable 5.1 demonstrates our ability to push frontier AI forward," said Taso Du Val, founder and CEO of Toptal.

"Developing frontier AI models requires more than just powerful compute. It requires specialized human expertise to train, evaluate, and continuously improve these models. ExpertCrowd brings together the technology and domain expertise required to build and train AI models and systems for executing complex, real-world tasks with precision."

About Toptal

Toptal is the world's largest talent network of highly skilled domain experts, connecting businesses with the top 3% of freelance talent on demand. Toptal was ranked the #1 Most Reliable Professional Services Company in America by Newsweek and Statista, based on an independent survey of more than 2,400 decision-makers at Fortune 500s. Founded in 2010, Toptal has served more than 35,000 clients across more than 140 countries. For more information, please visit Toptal.com.

Contact

Luiz Baia

SVP of Marketing Strategy

luiz.baia@toptal.com

Cision View original content to download multimedia:https://www.prnewswire.co.uk/news-releases/toptal-launches-expertcrowd-to-train-ai-with-vetted-domain-experts-releases-frontier-swe-benchmark-302907298.html