TechMiners ads

By the time a buyer starts Tech DD, a price is already on paper. A few weeks later, a report reaches the investment committee, and every finding in it comes with a cost: a price adjustment, an escrow, a specific indemnity or a W&I exclusion. Sellers rarely see that report before it is used in the negotiation. A Vendor Tech DD turns this around: you assess your own technology first, fix what can be fixed and decide how to present the rest. The issues are common enough to make this worthwhile. Black Duck’s 2026 OSSRA report found licence conflicts in 68% of 947 audited commercial codebases, and a buyer running the same scans will find them too. One of your CTOs, Florian Pahl, explains what to assess first, how to sort your findings and when to start. Read the full article.

Every Tech DD we run covers at least 99 maturity topics. We sorted them by one question: what decides the rating? AI vendors increasingly promise automated due diligence. So we checked where the evidence for each topic can come from, and where AI adds to it. • 40 topics can be rated largely on metrics; AI can support 15 • 38 combine metrics, questionnaires and interviews; AI can support 25 • 21 rest on expert judgement; AI can support 15 That makes 55 topics where AI can contribute evidence, and the share is highest where metrics run out. Language models read code, tickets, wiki pages and contracts at a scale no analyst can. They tell us, for example, whether a product is a thin wrapper around a third-party model or whether tickets carry real acceptance criteria. AI informs the rating; the judgement calls stay with people. Six topics, such as AI use in recruiting, depend almost entirely on interviews. Whether the architecture carries the next growth stage, or whether the team can deliver the plan behind the investment case, still comes out of conversations with the CTO and the engineers. Where a client or target rules out AI, as many NDAs now do, we deliver the full Tech DD without it. Our data analysis and our analysts' judgement do not depend on language models. Beyond this minimum scope, AI has paid off most in large and legacy code bases. We wrote up how we do it, and where a single prompt fails: https://lnkd.in/d3Mfupn3 #TechDueDiligence #PrivateEquity #AI

TechMiners conducted Tech Due Diligence for Upliift ahead of their majority stake investment in iWE, a low-code SaaS platform for insurance and real estate professionals that centralises data, automates workflows and connects third-party tools. Our metadata-driven assessment, powered by our proprietary TechMiners ONE platform, mapped iWE's software architecture, engineering collaboration, knowledge-sharing practices, cybersecurity posture and the scalability of its processes. Our experienced Senior Tech Analysts / CTOs turned the data into a clear, practical read on risk and opportunity. Congratulations to Jérémy Prats and the wider Upliift team on the investment, and to Paul-Marie RONGICONI and the team at iWE on this new chapter. Thanks to Upliift for trusting TechMiners, Constantin von Zitzewitz, Florian Pahl & Daniel Jung, with our signature, data-driven Tech DD. #TechDD #DealAnnouncement #LowCode

TechMiners conducted Tech Due Diligence on COPS GmbH (corima) for FLEX Capital and its portfolio company Omikron Systemhaus GmbH & Co. KG (multicash), ahead of the merger of the two businesses into an integrated end-to-end payments and treasury platform as a central building block of the CFO suite of tomorrow. Our metadata-driven assessment covered COPS's software architecture, engineering collaboration, knowledge-sharing practices, AI strategy and cybersecurity posture. We also looked into potential synergies between the two product and technology stacks, surfacing both functional overlap and complementary strengths, and advised on how to bring the teams, technology and processes together. Congratulations to Peter Kautz, Gary Kelling and the team at FLEX Capital, and to Dr. Daniel Appelhoff and the teams at Omikron and COPS. This is an exciting step toward a leading European platform for payments and treasury. Thanks to FLEX Capital for trusting TechMiners, Daniel Jung, Constantin von Zitzewitz & Mario Baumgart, with our signature, data-driven Tech DD. #TechDD #DealAnnouncement

TechMiners conducted a Red Flag Tech Due Diligence for 24Haymarket ahead of their seed investment in Mayday, a financial workflow automation platform helping finance teams automate month-end close, intercompany reconciliation, and multi-entity management on Xero and Quickbooks. Our metadata-driven assessment gave us a data-driven view into Mayday's codebase health, engineering collaboration, AI strategy, and security posture. Our analysts added the judgment a data scan alone can't provide. Congratulations to Jamie Dunnett and the team at 24Haymarket, and to James Scott-Griffin and the team at Mayday. Thanks to 24Haymarket for trusting TechMiners, Constantin von Zitzewitz, Val Bykanov & Christina Holderer, with our signature, data-driven Tech DD. #TechDD #DealAnnouncement #FinTech

Several deal teams told us this year they plan to run Tech DD by pasting a codebase into an LLM and asking for an assessment. Some have tried it. The outputs look like findings, but they are not. Three problems come up every time. Context windows compress exactly where the architectural risk lives. LLMs estimate rather than compute, we've seen vulnerability counts off by a factor of three compared to what a scanner returned on the same repository. And one broad prompt covering architecture, security, dependencies, and team structure doesn't do the work of five separate analyses. What we built instead: a multi-agent system that breaks a codebase into coherent units and runs specialist passes over each. A reviewer agent sits between stages as a quality gate, nothing reaches synthesis without passing review. Every finding carries a provenance record tied to a specific file and symbol. A claim that cannot point to a path and a symbol does not become a finding. Florian Pahl and Armin A. wrote it up. Find the full article on the TechMiners blog: https://lnkd.in/d4hw-Q8t