That burger is Lavern<\/a>, an open-source \u201cmulti-agent legal system\u201d that Innanen, a Finnish lawyer and law firm founder with a background in legal design, released this spring under an open-source Apache 2.0 license.<\/p>\n The \u201cmeat\u201d it mimics is a modern law firm \u2014 but this veggie burger version is staffed by a team of 67 AI \u201cspecialists\u201d that read your documents, argue with each other about what they find, and hand back the result in a typeset memo.<\/p>\n The agents come with names, faces, skill ratings and personalities. You assemble your team by flipping through what look very much like trading cards.<\/p>\n (Spoiler alert: There is even an Easter-egg agent named after a prominent legal AI endorsement deal. It is the most expensive lawyer on the roster but can do absolutely nothing.)<\/p>\n And yet, even with the whimsy, Lavern is a serious platform taking an alternative approach to how legal AI is built \u2014 and who gets to look under the hood.<\/p>\n \u201cIt works. It produces comparable results to any of the legal tech tools in the market,\u201d Innanen told me in a demo last week, running the system locally from his home in Alicante, Spain.<\/p>\n \u201cBut it\u2019s not all fun and games. I seriously think that the agentic debate can produce better and more varied results than just single agents. And that\u2019s what we are testing here.\u201d<\/p>\n Innanen\u2019s background helps explain why he built Lavern. He was the founding managing partner of Dottir Attorneys<\/a>, a Helsinki tech-and-media boutique, and went on to found Dot., a legal-design agency that spun out of the firm, and to co-found Legit, an AI company focused on helping organizations in regulated and high-stakes contexts adopt AI responsibly.<\/p>\n He helped start the Legal Design Summit and the Finnish Legal Tech Association. His initials, as his LinkedIn profile points out, are A.I.<\/p>\n Lavern is a project of his Legit and Dot. ventures, built over about six months.<\/p>\n At his former firm, Innanen had worked in mixed teams \u2013 lawyers alongside designers, technologists, and plain-language experts \u2013 and found the output was simply different, often better, than what all-lawyer teams produced. But no legal tech tools let him replicate that experience.<\/p>\n \u201cIf you think of almost all the legal tech tools in the market, the analog is the capable but very literal junior lawyer that you\u2019re working with,\u201d he said. Lavern, by contrast, is \u201cmore of a team than an assistant.\u201d<\/p>\n It is an attempt to recreate that dynamic with agents \u2013 not just one capable assistant, but a roundtable of distinct personalities debating a problem, with an orchestrating agent that mines the disagreement for the interesting angles and compiles them into a deliverable.<\/p>\n \u201cIf you think of four talented lawyers discussing a case, this tries to use that as the analogy,\u201d he said.<\/p>\n Lavern\u2019s interface is structured like an engagement, and it shows off both the substance and the quirk.<\/p>\n It opens on a landing page that reads \u201cYour firm is ready,\u201d with three budget tiers \u2014 Counsel (an expert opinion, up to about $10 in model costs), Review (a dedicated team, up to ~$40), and Full Bench (every specialist, up to ~$125). You can drop a document and go, or choose the recommended \u201cFull Engagement.\u201d<\/p>\n That begins with intake and briefing \u2014 which is, Innanen argues, where most legal tech tools fall short. \u201cMost of the legal tech tools don\u2019t really have any kind of intake system. The lawyer is the context.\u201d<\/p>\n Lavern offers two forms of intake. Either just \u201cDrop & Go\u201d a document or text, and it will figure out the rest. Or go through a guided interview that walks through questions such as jurisdiction, budget and fee structure, and potential conflicts. The interview is conducted by one of the AI \u201cpartners\u201d (or multiple partners if you prefer), who scope what you actually need before any work begins.<\/p>\n \u201cLaw is a context game,\u201d as the website puts it. \u201cThe model is not the bottleneck. Context is.\u201d<\/p>\n Next comes strategy, where you pick the approach you want Lavern to take in analyzing your issue. Options are:<\/p>\n These options differ in the number of steps they go through and human-approval gates they include.<\/p>\n You can also crank up the intensity so the system routes to a more capable, more expensive model, or dial it down and it drops to cheaper or local models.<\/p>\n \u201cTypically in legal tech tools you can\u2019t really choose the process,\u201d he noted, even though there is a big difference between reviewing an NDA and litigating a bet-the-company case.<\/p>\n Then come the playing cards for selecting your team. Each of Lavern\u2019s agents is rendered as a collectible card, complete with a skill radar chart, personality bars, practice areas, and a price.<\/p>\n Each of Lavern\u2019s agents is rendered as a collectible card, complete with a skill radar chart, personality bars, practice areas, and a price.<\/strong><\/p>\n<\/div>\n There are partners (Managing, Supervising, Of Counsel, a Risk Partner, a Litigation Partner) and a deep bench of specialists (M&A, regulatory, privacy, tax, restructuring, an IP specialist billed as \u201cThe Investor\u201d). You draft your lineup, manage your bench, and, when it gets unwieldy, do a little \u201cagentic HR\u201d and fire some.<\/p>\n You can also build your own agent from archetypes with names like The Shark, The Scholar, The Diplomat, and The Whitehat. If you want, you can clone yourself, or even an entire firm: paste a law firm\u2019s homepage, and Lavern reads the site and generates agents modeled on the real, named lawyers it finds.<\/p>\n During our demo, Innanen cloned his former firm, watching former colleagues materialize as agents. \u201cThis is a crazy feature,\u201d he admitted, \u201cbut I think it\u2019s worth exploring.\u201d<\/p>\n Once the work runs, you can sit by and watch the debate board. Agents post findings, each required to cite specific text from the document, challenging one another with counter-evidence, and an orchestrator resolves the disputes.<\/p>\n Unresolved or critical questions get flagged at a human gate for you to decide, unless you choose to let it \u201cwing it\u201d on autopilot.<\/p>\n The ultimate output is the sort of deliverable you would expect \u2013 a memo, a redline, a board email \u2013 with confidence scores, grounding indicators, and an audit trail, available in Traditional, Expert, or \u201cAccessible\u201d styles and in Word, HTML, or worksheet formats.<\/p>\n In my demo \u2013 a terms-of-service draft for a Delaware online-dating company \u2013 Innanen used a single agent to save money. It cost $2.52 of a $10 budget and took three minutes.<\/p>\n Lavern\u2019s engineering is where its real claim to originality lies, and the README on its GitHub page<\/a> lays it out.<\/p>\n Each agent, it says, is a specialized system prompt with its own role, MCP permissions, and slot in the debate protocol. All 67 run on the same underlying frontier LLM \u2014 Anthropic\u2019s Claude (the U.S. default) or Mistral (for EU data sovereignty) \u2014 or on a fully local model via Ollama.<\/p>\nA Team, Not An Assistant<\/strong><\/h3>\n
A Walk through the Firm<\/strong><\/h3>\n
\n
![\"\"](\"https:\/\/i0.wp.com\/www.lawnext.com\/wp-content\/uploads\/2026\/06\/02-partner-cards.png?resize=803%2C526&ssl=1\" "\\"\\"")
<\/a><\/p>\nAgents that Challenge Each Other<\/strong><\/h3>\n