Credal.ai

Credal.ai, founded in 2022 by former Palantir colleagues Jack Fischer and Ravin Thambapillai, operates as a secure AI agent platform for enterprises. The New York-based company was established to address the critical data security and governance challenges that arise when integrating large language models (LLMs) into corporate environments. The founders' backgrounds at Palantir, working on high-stakes data projects for the US Department of Defense and public health infrastructure, directly informed Credal's security-first approach. Thambapillai, the CEO, also has experience from Google, The Hut Group, and GoCardless, while Fischer's expertise includes airline reliability and defense computer vision.

The company's core business revolves around enabling enterprises to build and deploy secure, specialized AI agents that can collaborate on complex workflows. Credal’s platform acts as a secure data layer, connecting to an organization's internal data sources—like GSuite, Salesforce, and Slack—and allowing AI models to use this data while strictly enforcing the original user permissions. This solves a major pain point for companies wanting to leverage generative AI without risking exposure of sensitive information. The business model includes flexible pricing options such as annual contracts, pay-as-you-go plans, and on-premise private deployments to cater to different organizational needs. Credal serves a range of clients, from technology companies like MongoDB, Wise, and Checkr to large enterprises such as Comcast and the US Federal Government.

Credal's product is an end-to-end enterprise Retrieval-Augmented Generation (RAG) platform. Key features include the automatic syncing and inheritance of permissions from source systems, sensitive data redaction (PII masking), and granular audit logs to track data access and AI usage. The platform allows employees, even without developer skills, to build AI agents for specific tasks like prospect research, compliance checks, or answering HR questions. For developers, it offers a flexible API that is backward-compatible with models from OpenAI and Anthropic, simplifying secure integration. Credal supports both cloud and on-premise deployments and is SOC 2 Type 2 compliant. The company participated in Y Combinator's Winter 2023 batch and has secured $5.3 million in total funding, including a $4.8 million seed round led by Spark Capital.

Keywords: AI security, enterprise AI, data governance, large language models, AI agents, secure AI platform, retrieval-augmented generation, RAG, access controls, data compliance, Palantir alumni, Y Combinator, AI data privacy, sensitive data redaction, multi-agent workflows, enterprise search, AI audit logs, SOC 2 compliant AI, on-premise AI, enterprise chatbots