Find the software risks that could change your investment decision.
Before signing or committing capital, learn what the product depends on, what it will take to operate and where the roadmap could slip. Wavect reviews the available evidence and gives you a ranked report for your investment committee or board.
Technical due diligence is an independent review of the software behind a company or product before an investment, acquisition or board decision. It ends in a written risk report with severity, evidence and remediation options, not in legal or financial advice.
Best for
Investors, acquirers and boards who need an outside view of a software asset before a decision
Founders who want their own platform reviewed before investors ask the questions
Consider another approach
Legal, tax or financial due diligence, or a statement on valuation
Taking over and running the code after a provider change
Start with the decision the report has to support
An investor before a round, an acquirer before signing and a board reviewing its own platform need different depth. We agree the decision, the systems and repositories in scope, the access available and the recipient of the report before work starts. Commercial, legal, tax and financial diligence stay with your counsel and advisers. Where a technical finding touches their work, such as unclear code ownership or open-source licence obligations, the report flags it for them instead of drawing a legal conclusion.
What the review covers
Coverage follows the agreed scope. Each area records what we inspected, what we could not access and how confident the conclusion is.
Code and architecture
Structure, test coverage and quality, documentation, key-person dependencies and how hard the system is to change.
Security posture
Access to code and infrastructure, secret handling, dependency vulnerabilities and obvious exposure. This is a review, not a penetration test.
Delivery process and team
How changes reach production, release and rollback practice, incident history and who holds the knowledge.
Scalability, operations and cost
Hosting setup, monitoring, recovery readiness and the infrastructure and third-party costs behind the product.
AI-specific evidence
For AI products, evaluation design, model and prompt records, provider concentration, failure handling and task-level cost. Our [AI MVP due diligence checklist](/blog/technical-due-diligence-ai-mvp/) lists the evidence we look for.
Dependencies and tech debt
Outdated frameworks, unsupported components and the debt that will slow the roadmap, with a rough sense of what each item costs to address.
What could disrupt Day 1 after an acquisition?
We trace who controls hosting, identity, source code and key vendor accounts for the agreed product. If the seller shares systems or key people with another business, the report identifies the dependencies to resolve before handover, the risks to budget for and the questions still open.
A report that ranks risk instead of listing everything
Every finding has a severity, the evidence behind it, its likely impact on the business case and the realistic options to fix it. Where you ask for remediation effort, we give ranges, clearly marked as estimates under assumptions agreed for the engagement, not as quotes. The report separates deal-relevant risks from normal engineering housekeeping and lists the questions management still needs to answer. It does not predict valuation, deal terms or the outcome of a transaction.
How a due diligence runs
Scope, timeline and fixed price are agreed in a signed statement of work before work starts. Timelines depend on the size of the system and how quickly access and interviews are available.
Scoping and access
Agree the decision, systems, recipients and confidentiality. Receive read access to repositories, the data room and relevant documentation.
Review and interviews
Inspect the code and infrastructure, and talk to the people who build and run the system. Open questions go to management in writing.
Report and readout
Receive the written report and walk through it with your team. Factual corrections from the company under review are handled as agreed up front.
After the report, choose the right next step
Diligence ends with the report. If you then inherit the code or change provider, access recovery and stabilization are Software Takeover. Ongoing release testing is Software Quality Assurance. Adversarial testing and hardening of an AI feature belong with AI Security; the running cloud environment with Cloud Security. If the product was built with Lovable, Bolt or similar tools and needs fixing, see Vibe Coding Production Readiness. If the company needs a technical leader on its side through a raise or a sale, that is a Fractional CTO seat. If we may be asked to quote for remediation, we say so before the review starts.
Preparing your own company for diligence
Founders can commission the same review on their own platform before a raise. The goal is to find the gaps while there is still time to close or explain them. Useful preparation includes a dated architecture diagram, a dependency inventory, access ownership in company accounts, release and incident records and, for AI products, versioned evaluation results. If the product came out of an AI builder, read due diligence on a Lovable, Bolt or Replit app first.
// proof
Assessment work in existing engagements
None of these cases is a formal investor or M&A due diligence. Cedars documents a technical assessment with architecture and requirements records, targeted source review, a scoped AI review and prioritized actions with recheck criteria. Polity documents traceable analysis that identified and prioritized remediation across seven services. Soundromeda documents risk triage and hardening of a prototype before grant committees reviewed the build. They show the review method, not a diligence outcome.
▸These are selected projects, not our full portfolio. We have shipped 100+ products since 2018.
Questions before you start
We distinguish findings verified against records from management statements and open questions. The report explains which conclusions depend on missing evidence, what to request and what the gap means for your decision.
No. We review the software, its delivery and its operation. Contracts, IP assignments, licences, tax and financial statements stay with your counsel and financial advisers. Where a technical finding affects their work, the report flags it for them without a legal conclusion.
No. The report describes the technical risks we found within the agreed scope and the evidence behind them. How that affects price, terms or the decision to proceed is yours and your advisers’ call.
It depends on the size of the system, the depth you need and how quickly access is granted. After a short scoping call we agree scope, timeline and a fixed price in a signed statement of work.
Read access to the repositories, infrastructure configuration and the data room, plus time with the people who build and run the system. If some access is not granted, the report states which conclusions are limited because of it.
Yes, as a separate engagement after the report, and only if you want that. If there is any chance we will quote for remediation, we disclose it before the review starts so you can judge the independence of the findings.
It connects technical evidence to a consequence, such as delayed releases, operating cost or dependence on one person. Findings should distinguish confirmed facts from uncertainty and explain what remediation involves. Missing access must remain visible as a limit on the assessment.
Describe the decision, the company or system under review and your timeline. A short, non-confidential outline is enough for the first conversation; we sign an NDA before any data room or repository access.
Optional analytics, ad measurement, personalized Google ads, AI chat and embedded media load only with your consent. Privacy policy · How Google uses data
Ask our AI assistant
LeadWin by Feedbax and its AI providers process your messages and technical context, including outside the EEA. AI can make mistakes. Opening social content may also send data to its platform. Please do not share sensitive or confidential information.