Confidential

TechTruth.ai

AI Powered Audit Pitchdeck

Audit Report

NewCoEnergy

Energy AI Heavy Seed Stage MRR: €18K

8.4 /10

Verdict: Optimistic

WCCAA — Why Commit Capital
Auto-Weighted Average

Report IDWCC-2026-demo-1

Date09 March 2026

Model Training19 classification · 89 human review · 722 AI battle lessons

BenchmarkTop 10% of 41 Seed Energy decks

This is a fictitious sample report. The company "NewCoEnergy" does not exist. This report is for demonstration purposes only, to illustrate the structure and depth of a TechTruth.ai audit.

⚠ Not reviewed by human expert Fictitious sample — not a real company

This report is generated by TechTruth.ai using dual-model AI analysis (Gemini + Claude). Results are indicative only and do not constitute professional audit advice or investment advice. All findings should be independently verified. TechTruth.ai accepts no liability for decisions made based on this report.

TechTruth AI Generated Report

// [ANONYMISED DEMO] · Energy · AI Heavy · Seed · Report ID: WCC-2026-demo-1 · 09 March 2026

⚠ Not checked by human expert

Classification Energy AI Heavy Seed MRR: €18K

🏆 Top 10% of Seed Energy decks Benchmarked vs 41 Seed Energy decks · 19 classification · 89 human review · 722 AI battle lessons applied

WCCAA Score

8.4 /10

Verdict: Optimistic

1. Executive Scorecard (Table)

Dimension	Score (1–10)	Notes
Founder Team	9	Strong industrial GTM + engineering leadership; verification mix of supported / partial (see Founder Check).
AI Asset Depth	8	Proprietary TSDB + query plane — not a thin chat wrapper.
Technical Moat	8	Years of specialised time-series R&D create replication cost.
Infrastructure & Scalability	8	Architecture built for high-cardinality plant data.
Data Strategy	7	Customer-owned telemetry — strong privacy, limits pooled training.
Why Commit Capital Auto Weighted Average (WCCAA-score)	8.4/10	Weighted: Founder 70% · AI depth 10% · Moat 8% · Infra 6% · Data 6%

2. Executive Summary

NewCoEnergy is presented as an industrial analytics layer on top of a proprietary time-series stack — with AI used as a controlled query and insight layer over deterministic plant data. The principal Reality Gap in this fictitious demo remains execution timing: several roadmap milestones read as future “vision” even where the calendar has moved forward, and MLOps depth for autonomous forecasting is still thin in the materials. The updated founder verification block below reflects the same pipeline used in live reports: structured LinkedIn checks, open-web corroboration, optional exit traces, and a roadmap/experience fit pass — shown here with anonymised personas.

✓ Deep TSDB asset — differentiated vs generic LLM wrappers ✓ Founder verification matrix (Ghost Check) ⚠ Roadmap vs calendar alignment ⚠ MLOps / drift monitoring light in deck ✓ Early revenue signal (demo)

3. Founder Check (A first Ghost Check)

What was checked (this run)

LinkedIn pass 3 profiles structured scrape

Public web (DDG) Active founder corpus

Exit trace Limited public signals only

Roadmap × CV fit Cross-checked experience vs slides

Verification depth vs headline claim (sample)

Founder A · Commercial 61%

Founder B · Technical 58%

Founder C · Domain advisor 74%

Name & lane	Claim tested (deck, one line)	Evidence (Checked / Found)	Verdict	Avg verification
Founder A (CEO · Commercial)	“Scaled industrial software revenue across multi-region enterprise accounts.”	Checked: LinkedIn snippet, timeline. Found: Roles and tenure broadly align; deal scope wording softer than headline.	Partially supported	61%
Founder B (CTO · Technical)	“Ex–BigTech — led data platforms for large-scale telemetry workloads.”	Checked: LinkedIn, patent/news scan. Found: Strong infra credibility; limited public ML research trail vs claims.	Partially supported	58%
Founder C (Advisor · Domain)	“Former OEM operator — hands-on plant digitisation programmes.”	Checked: DDG, trade press. Found: Advisor capacity and sector background confirmed.	Supported	74%

Commercial ambition vs evidence

Founder A + C fit complex enterprise cycles typical of this category.

Technical ambition vs evidence

Founder B covers platform delivery; gaps remain vs stated autonomous forecasting path — consistent with ML/Ops hires still open in deck.

Founder list (anonymised): Three named personas above map to CEO / CTO / strategic advisor roles in the synthetic narrative — not real individuals.

Exit / sale verification: inconclusive for CEO headline exit (public corroboration partial).

Claim verification: Matrix reflects automated + human-assisted checks used in production reports.

4. Founder & Team Analysis

The team skews credible on delivery and industrial access; the verification table highlights where storytelling runs ahead of what open sources can prove.
Technical leadership is strongest on pipelines and uptime — lighter on published ML research than the “full autonomy” narrative implies.
Bus factor risk concentrates on engineering leadership for roadmap items that assume new ML hiring.
Synthetic demo only — conclusions illustrate layout, not an assessment of any real company.

5. The Main Pillars

Strategy & Product8/10

Infrastructure8/10

AI Logic7/10

ML & LLM Ops5/10

Team9/10

Data as an Asset7/10

6. AI Asset Deep-Dive (The Wrapper Check)

AI Class AI Heavy / Deep Asset

Proprietary model/stackConfirmed — TSDB core

Custom training dataMostly client-owned telemetry

LLM roleTranslation / orchestration — not sole intelligence

Technical moat replicabilityHard — years of domain-specific engineering

MLOps / drift monitoringEmerging — roadmap-heavy

7. Red-Team vs. Blue-Team Argumentation

Blue-Team (The Bull Case)

Industrial data is too messy for generic LLMs — whoever owns ingestion, canonical tagging, and low-latency query over live telemetry wins. NewCoEnergy’s specialised stack is exactly the kind of moat API wrappers cannot copy overnight.

Red-Team (The Bear Case)

Legacy SCADA naming chaos and integration debt cap how fast “AI insights” compound. Hiring for ML reliability lags the roadmap — execution risk concentrates on messy field deployments, not slides.

8. Meeting Focus List — Actionable Questions for Founders

Temporal Inconsistency & Traction

“Which roadmap milestones are live in production today versus still pilot-stage?”

Sensor Taxonomy & Cold Start

“How do you bootstrap models when tags are incomplete or inconsistent across plants?”

LLM Latency vs Plant SLAs

“Where is inference bounded vs deterministic queries — and what breaks if the LLM vendor changes pricing?”

MLOps for Forecasting

“Show the retraining, drift, and evaluation loop you will run per customer.”

9. The Why Commit Reality Check

NOTICE: Preliminary Vision Assessment. This synthetic page mirrors live report layout only. For a definitive Truth Gap on code, a GitHub Ghost Scan can be conducted — zero-knowledge metadata extraction for velocity, attribution, and dependencies — e.g. with Enjins.

Element	Founder Claim	Reality	Match
Founder / Team Claims	Executive scaling + BigTech platform leadership	Verification matrix: mixed — partial vs supported (see §3).	Gap
AI Architecture	“Natural-language access to industrial telemetry”	TSDB-first architecture; LLM assists query/explain — aligns.	Match
Performance	Sub-second queries on high-cardinality streams	Plausible with stack described; third-party benchmarks not attached.	Match
Roadmap	Autonomous forecasting milestone pack	Deck timing fuzzy vs stated calendar — treat as execution risk.	Gap
MLOps	Continuous learning “on plant data”	Monitoring / retrain loop under-specified in materials.	Gap
Data	Customer insights improve with every deployment	Strong privacy posture limits cross-customer training — flywheel nuanced.	Gap

Verdict: Optimistic (demo). Moat narrative is technically serious; founder claims require the same disciplined verification you see in §3–4 on real deals. Track record of shipping in messy plants matters more than slide superlatives.