NC-V · verification
Proof
An evidence dossier, not a portfolio. Every accuracy claim, product, and credential on this site traces to a public artifact someone other than me has signed off on — collected here so you can verify before a single email is exchanged.
SCORECARD · every number, one place
run 2026-04-27
three accuracy numbers, three different tests
- 96.8%
- retrieval accuracy on production traffic — how often retrieval surfaces the right source chunks for real users.
- 74%
- end-to-end pass rate across six public suites — the whole system answering correctly on deliberately hard, out-of-domain corpora. A stricter bar than production retrieval.
- 96%
- FinanceBench retrieval hit-rate — the same retrieval test, measured on one public benchmark specifically.
They measure different things, so they differ. All three — plus the 0/297 hallucinated citations — come from the same harness, with methodology and raw run artifacts published for audit.
NC-V.1 · things I don't control
Third-party verification
Evidence produced or held by neutral parties — public benchmark datasets, funding databases, knowledge graphs, and credential issuers. The strongest kind, because I can't edit it.
Public benchmarks
independent · auditable · peer-reviewed
- claim
- Accuracy is graded against public, peer-reviewed corpora — not internal demos.
- what this proves
- The numbers are graded on public datasets I did not create, with the scorecard and raw runs published for audit.
- evidence
- 221/297 (74%) aggregate pass rate across six suites, run 2026-04-27 (Cohere
rerank-3.5) - 0/297 hallucinated citations
- 96% (144/150) retrieval hit-rate on PatronusAI FinanceBench (NeurIPS 2023, arXiv:2311.11944)
- Suites include CUAD (legal, NeurIPS 2021), openFDA (drug labels), and SEC EDGAR 10-K filings
- 221/297 (74%) aggregate pass rate across six suites, run 2026-04-27 (Cohere
- verify
- BENCHMARKS.mdraw runsFinanceBenchfull breakdown
Funded venture
third-party record
- claim
- Co-founded SculptAI (a 4-agent game-dev pipeline, 2024–2025) which raised $350K in seed funding.
- what this proves
- A third party committed capital — recorded in a public funding database.
- verify
- Crunchbase: Definity LegendSculptAI case study
Knowledge-graph identity
third-party record
- claim
- The "Nic Chin" entity is registered on Wikidata — the same identity layer LLMs use to ground entity claims.
- what this proves
- This is a disambiguated public identity, not a name that could belong to anyone.
- verify
- Wikidata Q138698158LinkedIn
Vendor credentials
issuer public · identifier on request
- claim
- Credentials issued by IBM (AI Agents & RAG), Microsoft (Generative AI), and Google (Prompting), plus a University of Northampton degree.
- what this proves
- The issuers are independent and named; the issuer is verifiable.
- verify
- identifiers on request
Independent client reviews
platform-verified · on request · NDA-restricted
- claim
- Verified by Upwork as Top Rated Plus (top 3% of talent globally) with a 100% Job Success Score across multiple completed contracts.
- what this proves
- A neutral platform — not me — verified the review history and outcomes.
- verify
- shared in first call
NC-V.2 · things I built
First-party evidence
Work I produced — but shipped in a form you can inspect yourself: live products you can log into, open-source code you can read, and a design you can audit.
Live products you can use now
first-party · publicly usable
- claim
- Two production SaaS products built and shipped as sole architect — publicly accessible, no demo videos in place of a live URL.
- what this proves
- The work runs in production, not just in a slide deck.
- evidence
- sureciteai.com — multi-tenant document intelligence RAG
- systemaudit.dev — codebase intelligence reports in under 3 minutes
- verify
- sureciteai.comsystemaudit.dev
Open-source code
first-party · openly inspectable
- claim
- The benchmark harness, evaluation scripts, and product source for SureCiteAI are public and auditable.
- what this proves
- The implementation is inspectable — commit history, diffs, and raw artifacts included.
- verify
- github.com/nicukBENCHMARKS.mdSystemAudit engine (MIT)
A deterministic layer that overrides the model
first-party · design documented
- claim
- SystemAudit does not rely on LLM output alone — a deterministic pass re-checks every claim against measured facts and overrides the model where they disagree.
- what this proves
- A hallucinated finding cannot survive to the final report; correctness is governed, not trusted.
- verify
- the two-layer architecture in full
NC-V.3 · the person
Engineering track record
The products are proven above. This is the record of the person behind them.
Engineering track record
first-party · externally corroborated
- claim
- Production AI systems architect and fractional AI CTO.
- what this proves
- The person behind the products has a public, checkable history.
- evidence
- 13 production AI systems designed and shipped
- Two live SaaS products as sole architect (SureCiteAI, SystemAudit)
- Open-source AI evaluation framework with a public run history
- Fractional CTO engagements across funded startups
- Public technical writing and GitHub history
- verify
- About Nic ChinLinkedInGitHubPortfolio
NC-V.4 · the limits
What is deliberately not here
- Named client testimonials with company logos. Engagements run under NDA.
- Screenshots of internal client systems, dashboards, or proprietary data.
- Full certificate URLs, which expose the holder's legal name. Issuer is shown; identifier is shared on request.
why this page exists
Most portfolios ask you to trust the author. This page is designed so you don't have to.
Wherever possible, claims are backed by public benchmarks, live software, open-source code, or third-party records anyone can inspect independently. Where something can't be verified publicly — client confidentiality, personal privacy — I say why, and provide verification during procurement.