Generic tech DD templates miss what actually kills deals in 2026: AI claims that aren't production-grade, architecture that won't survive 2× traffic, and eng teams that look senior on LinkedIn but can't ship.

This checklist is what I use when doing independent engineering due diligence for VCs and acquirers ($1,250 fixed one-off, one to two weeks).

Architecture & scalability

• Can the current architecture handle 2× and 10× users without a rewrite?
• Monolith vs. services: was the split intentional or accidental?
• What is the actual bottleneck — database, queue, third-party API, or frontend?

Code health signals

• Test coverage: meaningful tests or coverage theater?
• Deployment frequency and rollback story
• Dependency risk (unmaintained packages, license issues)
• Incident patterns in the last 90 days

Team depth

• Bus factor on critical systems
• Seniority mix vs. what the pitch deck claims
• Hiring plan realism (can they actually close the roles?)

AI / ML claims (often skipped in generic DD)

• Production AI vs. demo-ware: evals, fallbacks, cost controls?
• Model routing and token spend — is there a bill shock coming?
• Data pipelines: where does training/inference data actually come from?

Security baseline

• Secrets handling, auth model, dependency vulnerabilities
• Compliance gaps that block enterprise sales

Deliverable investors actually use

Not a 200-page PDF. A traffic-light summary per area, specific findings with evidence, and recommended questions for management.

---

Full scope and pricing: https://themuneebh.com/technical-due-diligence

I'm Muneeb Hussain — Head of Engineering running production AI systems, fractional advisor for founders on the side.