Galileo Research

A category-wide strategy & architecture map of the enterprise AI deployment layer — 20 companies (independents, Palantir, the labs' arms, and the incumbent SIs) scored on a 5-axis master matrix, organized by strategic axis rather than company. The load-bearing finding: the "ontology-first (Palantir) vs. workflow-first (Distyl)" fork is quietly collapsing — Distyl now ships a persistent "Context Mesh" semantic layer while Palantir sells 1–5-day workflow bootcamps; the surviving daylight is the depth of the deterministic write-back substrate. The evidence spine is radical honesty: across the whole category, zero cleanly customer-triangulated outcomes exist (the GE/Palantir best case corroborates +26% but the 10-K never names Palantir); the viral "95% of pilots fail" stat is autopsied as a low-rigor preliminary deck; the incumbents are graded channel-not-roadkill (Accenture $5.9B GenAI bookings; IBM >$12.5B book). Every metric carries a CONFIRMED/REPORTED/ESTIMATED/CONSENSUS confidence label. 40 sourced citations, Chart.js matrix visualization.

A node-level DCF and payback surface for Star Catcher's orbital power-beaming utility, built as a gut-check before the full report. Explicit efficiency chain (collection → wavelength → beam/pointing → receiver capture), utilization haircut, and capex scenarios collide against two willingness-to-pay regimes. Core finding: array-avoidance pricing never pays back at any capex; the node only closes as a 2–4 yr infra asset under mission-enablement pricing ($2,500–5,000/delivered-W-yr). Breakeven WTP for a base node is ~$3,700/W-yr (~$1.42/kWh) — the single number the whole thesis reduces to. Two loud unknowns flagged estimate-not-verified: signed-PPA $/kWh and measured wall-plug efficiency. 11 tier-1 sources, 2 Chart.js visualizations.

A founder's-eye teardown of how AI forward-deployed engineering (FDE) deployment startups actually operate, anchored on Distyl and Northslope. How they win enterprise accounts, the deploy-then-productize motion, the frozen-model + harness architecture (vs fine-tuning), team and margin structure, the competitive ladder from seed through Series B, and where the durable moat — and the trap — sits. Framed for evaluating whether to found one. 30 sources, 2 Chart.js visualizations.

A fully re-validated rebuild of the Quiet Alpha Map. Every firm-to-thesis attribution was re-checked by reading the underlying claim (not keyword matching), correcting ~27% false positives, and 280 previously-untagged claims were mined — recovering 133 lone signals. Convergence bands recalibrated to the genuine distribution (Consensus 21+ down to curated Lone Signals), themes restructured (incumbents split into incumbents-vs-insurgents, the agent economy & trust layer merged, a new Space/orbital frontier theme added), citations rewired to the exact source essay, Sharp Contradictions expanded to 7, plus a 10-bet Lone Signals tier and a 5-bet "Off the Deep End" spotlight. 254 source-linked citations; includes a methodology changelog documenting the audit.

A synthesis of 906 forward-looking market claims from 80 leading venture firms, sorted by how many firms converge on each thesis — from priced-in Consensus down to Lone Signals. The crowded top shows what's already in the water; the sparse bottom is where edge hides. Includes Sharp Contradictions (where serious firms directly disagree) and an "Off the Deep End" section. 164 source-linked citations across 68 firms.

A founder's operating teardown of the embed-and-build model — anchored on Distyl ($1.8B) and Northslope. The core loop (embed FDEs → harvest proprietary context → build agentic logic → bill on outcomes), engagement lifecycle, team shape & the Palantir "founder factory," the Distillery architecture + verification layer, unit economics, six granular automation case studies (prior-auth, fraud detection, supply-chain RCA, contracts, order resolution, tax), and a 7-company peer field (Sierra, Harvey, 8090, Cresta, Cohere, Glean, Cognition, Decagon). Where the model commoditizes and the three open lanes for a new entrant. 36 sourced citations, 2 Chart.js visualizations.

Mapping the $725B infrastructure buildout reshaping the global energy grid. Hyperscaler capex profiles, the "Prediction Gap" (Goldman Sachs revised 2030 forecast upward 3× in 14 months: +165% → +175% → +220%), three-scenario forward extrapolation through 2031, RSI wildcard (Jack Clark: 60% odds by 2028), grid bottlenecks (160-week transformer lead times), and the Jevons loop. 50+ tier-1 sources, 8 Chart.js visualizations.

AI is to intelligence what coal was to energy. As the cost of thinking drops 280x in two years, Jevons' Paradox predicts demand will increase far beyond substitution. New categories, historical parallels, and investable implications for a world where intelligence is essentially free.

The largest US grid operator just admitted the power system is structurally broken. 6,517 MW capacity deficit, 60-80% cost escalation, construction timelines doubled. 12 investable angles mapped to specific structural problems. Second/third-order effects traced.

Understanding Hive's technology, business model, and brand philosophy — and what the "hive" concept means for a seed fund's positioning. Company overview, product suite (21 APIs), market position ($3.9B content moderation AI market), and Scarce Assets framework analysis.

Can AI agents that "join your team" like real employees create a venture-scale business in the $327B BPO market? Investment evaluation for Velanir (fka Botero Labs). 10 competitors mapped, BPO disruption analysis, and the coworker-vs-tool distinction.

Is the "Hollywood agent for tech talent" model viable at venture scale? Competitive landscape analysis for Standout (YC S26). 12 players mapped, precedent analysis, and the frequency problem that kills agent economics.

The fastest-growing enterprise AI company in history is approaching a $1 trillion valuation. Complete one-pager with funding history, revenue trajectory, competitive landscape, notable quotes, and bull/bear thesis. With Chart.js visualizations.

The US wine industry is in structural downturn. But brand value is separating from hard assets — and smart buyers are acquiring legacy brands for pennies on the dollar, without touching a single vine.

We scanned 181 academic categories. The 10 fastest-accelerating fields reveal three clear investment signals — and two traps where the money arrived before the science matured.

90% of manufacturing runs on PLCs. AI can't read the code. One startup is building the Rosetta Stone for industrial automation — investor diligence report.

The math is brutal, the base rates are terrifying, and most of the best investments looked insane at entry. A six-filter framework for finding 1000x outcomes at seed stage.

How contrarian conviction, extreme concentration, and "definite optimism" produced the best venture returns of a generation — and what a seed-stage AI fund can learn from it.

$258.7B of AI venture capital is chasing disruption — but it's concentrated in the wrong places. Gap analysis of 12 industries by AI vulnerability vs. founder activity.

Analysis of autonomous AI agent security vulnerabilities based on "Agents of Chaos" research. Maps the emerging agent security market and seed-stage investment opportunities.