Legora, founded in Stockholm in 2023, is building an agent-driven operating system for legal work that uses OpenAI models and coding agents to automate repetitive tasks such as contract and NDA review. The company applies the same AI tools internally that it offers to clients and reports measurable outcomes including major time savings, faster contract turnaround, and multi‑million dollar reductions in external legal spend.
Mission and origins
Legora’s stated goal is to embed autonomous, agentic capabilities into everyday legal workflows — what CEO and co‑founder Max Junestrand describes as “the agentic operating system for legal work.” The company started from the observation that legal teams are frequently overloaded, leaving important work undone. The release of GPT‑3.5 made many of the founding team’s experimental legal use cases feasible.
Technology and internal use
Legora integrates OpenAI models (including ChatGPT Work and Codex) and coding agents. These technologies enable:
- large‑scale document review and agentic legal workflows,
- parallel prototyping of many engineering approaches,
- rapid diagnostics and fixes by searching millions of lines of logs, sometimes taking a bug report to a working fix in about five minutes.
OpenAI’s models are both part of Legora’s product and shape the company’s internal operations; Junestrand emphasizes the importance of remaining AI‑native.
Engineering efficiency and product impact
According to CTO Jacob Lauritzen, coding agents let engineers prototype dozens of solutions simultaneously and select the best option in days. The same approach speeds incident response: where previously multiple engineers might investigate for hours, agents can pursue many hypotheses in parallel. While customers do not directly see engineering workflows, they benefit from faster issue resolution, earlier product feedback loops, and quicker testing with customers.
Legal practice changes and quantifiable customer gains
Legora’s General Counsel Emma Kadri Bergström describes how the company automated initial reviews of incoming NDAs using a playbook built from historical agreements. Before automation, NDA review took roughly 20% of the legal team’s capacity; after the workflow was introduced, it dropped to under 1% — a reduction of more than 95%.
Measured customer outcomes include:
- 90% faster contract turnaround for one Fortune 100 customer,
- more than $20 million saved on external litigation and diligence for that same customer,
- 9.5 hours returned per lawyer each week for higher‑value work.
Legora’s Tabular Review workflow can analyse thousands of contracts across multiple questions in minutes, letting lawyers apply detailed review far more broadly than traditional manual sampling allows. This increases the likelihood that contractual risks outside a small manually reviewed subset will be surfaced for lawyer assessment.
Implementation lessons
Legora highlights several practical lessons for adopting AI in legal teams:
- Lead by example: when over 75% of leaders actively use the platform, median usage across the organisation is three times higher.
- Start from a clear problem: focus on users’ goals, then identify where AI improves the workflow.
- Build internal champions to convert experimentation into sustained adoption.
- Be prepared to rebuild: model improvements can rapidly obsolete prior product assumptions.
- Design for trust: incorporate evaluation, testing and transparency from the outset.
The next frontier: long‑running agents and real‑time collaboration
Legora’s roadmap looks beyond making current workflows faster. Jacob Lauritzen expects two parallel advances: agents capable of longer, more autonomous tasks that require extensive context (for example, due diligence across millions of pages completed overnight), and faster models that enable real‑time, interactive collaboration between lawyers and AI rather than a simple task‑and‑response model.
Conclusion
Legora aims not only to improve the speed of existing legal tasks but to expand the scope of what legal teams can do: taking on work that was once impossible due to capacity limits and freeing lawyers for judgment, strategy and other high‑value activities. As Max Junestrand puts it, the company believes this is “the best time in history to be a lawyer.”
