The shape of the work.
When you need gaps filled, problems solved, and the work executed — not high-priced PowerPoint presentations, confirmation bias, and a billable-hour meter.
Client work stays confidential; structure doesn't have to. The seventeen exhibits below are representative schematics — real architecture and real scale, with every name and identifier stripped; figures are from the work itself or marked illustrative. The structural counts are genuine, taken from the artifacts themselves. (The following exhibits are just graphic generalizations of a small sample of work and do not capture total complexity or analysis.)
A success asserted by those paid to assert it.
A multi-billion-revenue operator ran an international program for one of the largest technology companies in the world, a U.S.-based Fortune 10 partner: distribution across the U.S., the U.K., the EU and Oceania, multiple currencies, regional engines, a Middle-East redistribution hub, and resale lanes touching every region of the world. Its success was conjecture — asserted by the people incentivized to make it look like one; the numbers, the profitability, the truth had never been validated. The answer didn't exist until it was built — so Jonathan built it, across every market: one model taking every region, currency and resale lane down to a single answer. It found taxes being overpaid, a program that wasn't profitable, and data nobody could tie out. He fixed it himself — the dashboards, the data, the program and every SLA — and rebuilt the partnership into the company's highest-margin line of business, with a 15–20%+ margin floor across EMEA and APAC.
THE SUBSTRATE IS NOT DECORATION — EVERY MARK BEHIND THE SCHEMATIC IS ONE OF THE 165 TABS, EVERY FAINT ARC ONE OF 4,545 CROSS-SHEET LINKS, DRAWN AT REDUCED DENSITY SO THE FLOW STAYS READABLE. THE SCENARIO BANK IS A STACK BECAUSE IT IS 43 CLONED ENGINES. AS THE MODEL COMES TOGETHER, THE DRIVERS BEHIND IT — LABOR, SHIPPING, DUTY, DEMAND ROUTING — LIGHT UP AND CONNECT.
LEAK LEDGER — SEVEN LANES BILLED WRONG AT ONCE, AMONG THEM: A STATUTORY REMITTANCE PAID TWICE · LABOR OVERPAID · CARRIER RATE INCREASES NEVER REPRICED · SLA ITEMS BILLABLE BUT NEVER BILLED · FIFO NON-COMPLIANCE QUIETLY DRAWING PENALTIES — EACH CALLED "INCONSEQUENTIAL." NOBODY HAD RUN THE MATH. TOGETHER: MILLIONS.
FORMULA CENSUS — 20,648 SUM · 11,293 OFFSET · 5,587 IFERROR · 1,914 ROUNDUP (headcount and kits always round up) · 0 NAMED RANGES
As found: regional engines invoicing in multiple currencies — one cut of a program that sold worldwide — a settlement chain nobody had mapped, dual in-/out-of-SLA revenue ledgers, and seven lanes being billed wrong at once, led by a statutory payment remitted twice: an error that had survived committed local staff and an in-region outside audit-and-accounting consultancy, caught only by translating the source documents line by line. Every lane had been called inconsequential, because nobody had run the math. Together they were millions. As rebuilt: one machine from partner sell-through, through the Mid-East hub and worldwide resale lanes, to consolidated USD cash — anchored to an invoice-level actuals export. The internal read was "they'll never go for these terms." The model said the counterparty already knew the program was losing millions — and was counting on that fear. No bluff, no movement: they conceded on every lane, and the restructured economics carried the program into five new markets. In plain terms, it found taxes paid twice, a program that wasn't profitable, staffing that was wrong — temps where permanent staff kept through the seasonal lows cost less — and production that wasn't optimized. Jonathan redesigned the program and every SLA, set up the dashboards, and tied out the bad data across every market; the model kept growing, to 198 sheets and 65,125 formula cells by its fifteenth version. Structural counts in the chips are real — taken from the workbook itself.
The complexity was the leverage.
A sub-investment-grade infrastructure asset, just out of bankruptcy, carrying syndicated bank debt under a credit agreement dozens of amendments deep — bolt-on language on bolt-on language that no one inside could fully model. Rather than simplify it, he used it: mapped every definition and cross-reference into one model, wrote most of the technical legal language himself, and routed the changes through definitions most readers treat as non-headline. A bulge-bracket syndicate working to a quarter-end close passed it through as written: coverage requirement eliminated, bank debt kept. Around it, the refinancing, the cash turnaround, and a new asset beside it.
Complexity is only a liability to whoever can’t model it. Jonathan was called in as the asset came out of bankruptcy, and built the plan that carried it. The credit agreement had been amended so many times that no one inside the company could model it, and the coverage-breach risk was building underneath. Jonathan mapped every definition and cross-reference into one model, wrote most of the technical legal language himself, and ran the changes through the definitions most readers treat as non-headline — the exclusions inside CFADS and the accounts that feed them. As first drafted, costs paid from an equity-funded liquidity reserve still counted against coverage; counsel’s fixes didn’t hold, and one cited an account the agreement doesn’t have. The bulge-bracket syndicate, working to a quarter-end close, passed the final language through as written. Eliminating the coverage requirement on a distressed, sub-investment-grade asset is close to unheard of — and the syndicated bank debt stayed. Around it, the capex and O&M curves were rebuilt from actual spend and an FP&A function was built from scratch, adding about fourteen months of runway and $250M+ of enterprise value. Jonathan’s design for the travel center, structured for optionality, won a $22.9M federal grant; the city approved it, and it is moving forward. Names, counterparties, and identifiers are stripped; structural counts come from the working documents. Amendment layers, coverage values, and runway line shapes are illustrative.
Sold what others dismissed — and kept what the business needed.
A public utility’s tower and communications sites. Internal studies with outside experts, repeated over about five years, had put the ceiling at about $5M — nobody, they said, would want the assets. Jonathan didn’t believe it, and people in the field told him he was wrong. He ran the process anyway: about $100M, roughly 29× cash flow — while the utility kept its equipment on the sites rent-free and an O&M contract back for the sites the buyer can’t even enter.
Ninety-two candidate sites, every one hunted down, listed, mapped, and deeded for the data room: 14 excluded, 53 offered for sale, 25 under a lease structure, and 77 closed. Even the sites inside secured substations sold. The buyer can't set foot on them, so it pays the utility to run their O&M, along with sites that touch federal restrictions. Those carried a lower multiple, but they still sold, and they had produced very little cash flow anyway.
The experts said $5M. Internal studies over about five years had concluded the most anyone would pay was about $5M, and people in the field told Jonathan he was wrong to try. He built the case on tower cash flow from the colocation leases — cash-flow multiples, with tenant-removal risk priced in — after hunting down, listing, mapping, and getting deeds to every tower and building the data room. Each site was triaged: sale or lease structure, inside a substation or not, critical to the backbone or not, assignable or not. A competitive bidder process took it to about $100M, roughly 29× cash flow. The closing critical path ran easement grants and assignments, partial master-lease assignments, tenant-lease amendments, surveys and fencing, and a transition-services agreement. What the utility kept mattered as much as the price: rights to all of its equipment on the sites, rent-free, won in the give-back calculation; access to its secured areas; and long-term O&M contracts back, with escalators, on the sites the buyer can’t enter. Dot placement is illustrative; counts come from the working site list; outcome figures match the home page.
Every side of a P3.
Few people have worked public-private partnerships from every seat at the table. Jonathan has: the state’s negotiating side of a private toll concession, procurement and finance at a state DOT, local partnerships and credit built on traffic-and-revenue studies, the selling side of a utility asset sale, and the counterparty’s side, bringing that same concession out of bankruptcy years later. Behind nearly every deal sits a long-term feasibility study.
One concession, three sides of the table. From 2009, on the state’s negotiating side of a private toll concession, working with senior state leadership and the legislature's finance committee: the non-compete that barred expanding the parallel interstate was struck, the concessionaire had to share certain impacts, and liquidated damages for breaches were written in, and they still bind the deal. The state's side won: the concessionaire overpaid on a hockey-stick traffic forecast that Jonathan said at the time could never make money, and the state came away with the money for other projects. When the forecast failed, the loss was the concessionaire's; years later Jonathan was called in from the other side to bring the concession out of bankruptcy (Exhibit B). A win from every seat. Around it: multi-billion-dollar P3 and toll procurements at a state DOT, local partnerships that got roads built, credit facilities sized on traffic-and-revenue studies, a reserve structure that carried one year’s excess revenue into later years inside public-finance yield restrictions, and the selling side of a utility tower sale run within the DOT’s guidelines (Exhibit C). The policy-side terms and the credit and reserve structures are Jonathan’s account; names, agencies, and counterparties are stripped.
Two roads, one balance sheet.
For a consortium client, Jonathan valued two connected toll corridors, then the combination. One is held by a public authority under a multi-lien bond indenture. The other is privately owned, with its cash trapped behind lock-up tests. Modeled on their own documents and put under a new debt structure, the two combined clear the coverage tests four years sooner — an answer that only shows up when every side's documents are modeled at once.
See every side before you price it. A consortium client wanted to know what two connected toll corridors were worth, alone and together. Jonathan modeled both from their own documents: the public corridor's master indenture, with four lien levels and their additional-bonds tests, its reserves and its capital spending; and the private corridor's lock-up tests, early-redemption schedule, reserves and debt-service reserve, with growth inferred from the owner's own impairment charges. The same public-corridor cash flows came out worth about twice as much to the public owner at its cost of capital as to a private buyer at its own. Combined under a new debt structure, with the same senior coverage tests, the lock-up cleared in 2023 instead of 2027. That freed four years of trapped cash for the private owner, which is the kind of value only visible when every party's documents sit in one model. The same structuring muscle runs through long-term power purchase agreements and P3 offtakes with power companies. All names, places and dollar values are withheld; ratios, years and model counts are from the client workbook.
Run the numbers yourself.
Twenty-plus capital- and debt-markets transactions with bulge-bracket banks and private credit, every one closed. On each one Jonathan ran the models independently, caught structural issues the banks missed, and designed structures bankers went on to use. Four views: the call option the market wasn’t pricing, a $255M reservoir financing, the rating agencies, and the critical path of a bond sale.
Never take the banker’s math on faith. Every capital-markets deal was a collective effort, and Jonathan ran the numbers himself anyway — which is how he caught structural issues the banks missed and why bankers ended up using his structures. On one major financing he reframed the decision around lifetime economics instead of total interest cost and headline savings, found the market underpricing the call option, tested it with trading desks, and drove the deal to the structure that saved millions while keeping the option to refinance. On a $255M reservoir, the debt structure, capex, and O&M were redesigned together: about $56M saved, about $83M of cash freed in four years, rates stabilized, and state funding secured. And after he challenged the rating agencies’ prevailing treatment of complex infrastructure obligations, a major agency invited him to help reassess its methodology. Advisors and underwriters were chosen by competitive bake-off, on merit. Call-option bars are illustrative; names and counterparties are stripped.
Trust is collateral.
A damaged private-credit relationship. The lender had found inaccurate information coming from inside the company and conduct that didn’t match the cash terms of the agreement, and the asset-based borrowing base — the dial that sets how much the company can draw — was broken. Jonathan took it over: rebuilt the data, the checks, and the models; changed how orders were filled so stale inventory stopped aging out of eligibility; and personally ran the borrowing base several times a week for 18+ months, the only person the lender would let touch it.
Take ownership of the mechanism that controls the cash. An asset-based line lives on its borrowing base: eligible receivables and eligible inventory, each at an advance rate, less the lender’s reserves. Here the data behind it couldn’t be trusted, so the base and the relationship were both failing. Jonathan rebuilt the data, validation checks, and models across the IT, finance, and warehouse sides; showed operations and executives how filling orders differently changed the cash position and return on capital, because inventory only counted toward the base for 180 days; and then ran the base himself, several times a week for more than eighteen months. Trust came back transaction by transaction: the lender extended better interest terms and loosened covenants in key areas, more room than the financials alone would have earned. Bar heights are illustrative; the collateral-base gain is his estimate of roughly 20–25%.
Data, weather, resources, demand, contingencies, observations in. Tradeable forecast out. Every night.
A production forecasting and trading-research system for wholesale power — weather, outages, and network topology streamed through 113 pipeline stages, five model classes, a battery of hardened statistics, and a falsification layer whose entire job is to kill its own ideas. Built and operated solo.
THE FALSIFICATION LEDGER — 192 CLIMATE CELLS TESTED, 37 SURVIVE FDR CONTROL · 42 SIGNALS SCREENED ACROSS 1,108 NODES AND 1.79M NODE-DAYS, 0 PROMOTED · 176,281 EVENT-ADJACENCY TESTS, 0 VALIDATED — THE NAIVE NULL WOULD HAVE SHIPPED 8,500 FALSE FLAGS, AND THAT COLUMN IS KEPT ON THE ARTIFACT RATHER THAN DELETED.
SELF-INCRIMINATION BY DESIGN — ONE FORWARD PRICE MODEL SHIPS WITH TRUSTWORTHY = FALSE, PUBLISHING ITS OWN LOSING SCORE (MAE 9.41 VS PERSISTENCE 9.30) INSTEAD OF BEING TUNED UNTIL IT WINS · IT IS SCORED AGAINST THREE NAIVE BASELINES, ONE DELIBERATELY GIVEN MORE HISTORY THAN THE MODEL GETS · A FOURTH COMPARATOR IS LABELED IN THE ARTIFACT ITSELF AS “NOT A FAIR COMPETITOR” SO NOBODY CAN QUOTE IT.
THE STACK — DuckDB · POLARS · PARQUET LAKE · DAGSTER · GPU-LOCAL INFERENCE · WINDOWS SCHEDULER + WATCHDOG · 35-TAB DASH DESK · 78,337 LINES · 1,348 TESTS. THE STATISTICS — ROBUST Z (MEDIAN/MAD) · BENJAMINI–HOCHBERG FDR · PERMUTATION · SPLIT-HALF · WALK-FORWARD · SUPERPOSED-EPOCH · CUSUM REGIME BREAKS · CONFORMAL PREDICTION BANDS · QUANTILE REGRESSION.
Scale is the easy part. Six source lanes across 43 endpoints — market and settlement, transmission and network topology, generation and outages, fuel, weather and climate, entity registry — land in a 263-million-row lake with 5.6 years of hourly nodal history and 85 years of climate. A hundred and five stages rebuild it nightly, unattended, behind a watchdog that re-runs the night if the freshness manifest goes stale. Rigor is the hard part. Five model classes compete on identical folds; promotion is earned by a stated skill threshold against naive baselines, never by a person’s approval — and the falsification layer exists to destroy candidates, which is why its headline numbers are zeros. Every dollar claim reconciles against the market operator’s own settlement records: rank-based validation had passed three P&L bugs that only the dollar oracle caught. Architecture and scale shown; no signal, threshold, node, or position disclosed. Built on public market data.
Ten thousand buses. Twelve thousand branches. The two hundred wires that matter.
A DC power-flow model of the ERCOT network rebuilt from the auction's own PSS/E case — exact shift factors for every bus, 11,016 element-by-contingency pairs that have actually bound, the stability limits that move by announcement — with a 2,176-asset generation fleet, a 1,789-project interconnection queue, and 59,563 unit outages pinned to the nodes they touch. Then one identity turns all of it into the value of any path — traded or not.
There are over a million possible paths; the engine never models them. It models the wires that bind — binding frequency times severity, per family, per month, per time-of-use — and reads every node's exposure to each wire straight off the network model. Value is additive across constraints, so a path nobody has ever traded is priced exactly like one that trades every month. Contingencies, stability limits, the fleet, the queue, and the outage book all live in the same coordinate system: the grid as it physically is, not as the price history remembers it.
What moves a wire, measured. What a path will pay, forecast. What it costs to win, priced.
Five weather regions and 85 years of climate, a 24-model ENSO plume, a seasonal load U-curve, gas at three hubs, the merit order of 992 plants, 5-minute grid telemetry — 127 covariates, each stamped with the day it became knowable. Two learned layers forecast the price at every node. A third prices the auction itself: the odds a bid clears, the winner's curse, and the 15.7% of paths the market pays you to hold.
The forecast is two layers: a physics prior counted from what the grid actually did, and a gradient-boosted correction that learns each node's basis and is graded on ranking, with uncertainty bands that are conformally calibrated rather than assumed. The auction layer is where most buyers lose — you pay the clearing price, not your bid, and the market runs about 1.7 losing trades per winner — so the engine prices the probability of clearing, the winner's curse, and the deration cap before a number reaches the ticket. What it never does is pull the trigger: bid submission sits outside every autonomy tier, permanently human.
What a large-load deal has to know before it signs.
The power-market systems behind Exhibits H through J, pointed at a transaction. Before a lease or a PPA is signed, four questions decide it: when the site really energizes, what congestion and curtailment it’s exposed to, what power will cost, and how reliable supply is. The market’s own data answers them.
Production power-market systems answer the deal's questions from the market's own data: an interconnection queue of 1,789 projects and about 439 GW, of which about 113 GW is signed to come online in 2026–28; transmission constraints and contingencies; the mapped generation fleet; and a detector that spots new large loads switching on. The same discipline as the deals: model it before you sign it.
The bridge between the power work and the deal work. The queue, constraint, contingency, and fleet counts come from the research engine’s own artifacts (October 2026); the large-load detector runs in the separate live market monitor. Everything is built on public market data; no signals, nodes, thresholds, or positions are shown. “Other” is the queue’s own fuel category. The questions are the ones a lease, PPA, or site acquisition has to answer before signature.
Acquired broken. Rebuilt whole.
A venture-backed company acquired with broken financials — and the cash had run out. Tracing it, Jonathan found orders being sent back because of inventory. From there, production, pricing, cash, and valuation were rebuilt from the ground up, down to the week each physical unit spends on the line.
FORMULA CENSUS — 1,567 SUMIFS bucket the cohorts · 825 MIN / 678 MAX clamp the ramps · 314 MONTH run the dates · 26 XNPV/XIRR sit at the top of the pyramid
It started in the red: the cash had run out, and tracing it turned up orders being sent back because of inventory. Unit-by-unit, week-by-week: ~130 physical units each get a row in a weekly build plan — the fifth iteration of the sequencing; earlier versions are archived in the book. Hospitality economics fused to a factory cost stack, a four-stage cash conveyor where money leaves months before revenue exists, seven turnaround levers wired as switches, and a valuation not allowed to mean anything until four independent lenses agree. No fragile circularity anywhere: dated cash flows and one-way reference lanes — circular only where it makes a point.
One question in. The right instrument out.
Not a menu — a working arsenal. Statistical inference, pricing studies, market and marketing science, decision trees, risk matrices, cost-benefit — the question picks the instrument, and every answer is tied back to authoritative figures before anyone acts on it.
The instrument is chosen by the question — an elasticity study for the price, an expected-value tree for the deal, a 5×5 matrix for the risk, a cost-benefit for the spend — instruments designed from zero and validated to survive an adversarial read. And every number that reaches a decision is reconciled first: key-figure tie-out checks run automatically inside the pipelines, so the answer has already survived its own audit.
Integrated multi-entity forecast.
Linked-sheet architecture reconciling par, interest, and defeasance across operating and corporate entities — under three board-adopted scenario cases.
Architecture only — no figures, entities, or client identifiers shown. This is a representative schematic of the model class, not an actual deliverable.
Resource need to project-finance viability.
Thirty-five-plus candidate financing scenarios, each overlaid on the issuer’s existing debt service. Revenue and operating cost are fixed; O&M is paid first out of revenue, and existing debt, new debt service, and lifecycle capex all draw on the CFADS that remains. Coverage stays red and unbankable until the new debt is sculpted and a reserve pre-funds the capex — then it clears the covenant, green across every period.
A project-finance cash waterfall, by year. O&M is paid first out of revenue, leaving CFADS. On top of the issuer’s existing debt service, each of 35+ candidate structures overlays its new debt service by year — and lifecycle capex lands in lumps from year 5 in differing amounts. When existing debt, new debt, and capex together outrun CFADS, coverage breaks. Base sets new debt flat and heavy and breaks in the ramp and every capex year; re-tenored sculpts it so the ramp clears, but capex still competes for the same cash; aligned sculpts the new debt and pre-funds a maintenance reserve, so DSCR clears its covenant every period. The three shown are representative of the full scenario set. Curves are illustrative shapes, not client figures.
Where the raise began.
The first tabs of the acquisition book aren't a model — they're a market argument. Four benchmark lenses anchor every assumption to the outside world, and the wedge they expose — hotel-grade rates on factory-grade cost per key — became the genesis of the investment raise.
Four benchmark sheets sit in front of the model — nightly rate against ~8 brands, build cost per key against every hotel category, operating margins, and a 26-line peer-ratio panel the SG&A glide path converges toward. They exist to make one argument: charge like a hotel, build like a factory. That wedge — anchored outside, then proven inside the 65-month engine — was the genesis of the investment raise. Bar heights are illustrative shapes, not client figures.
The rigorous approach to marketing studies and pricing strategy.
Conjoint analysis, pricing strategy, question design, targeting, and the statistics underneath them, in two cases on one layout. Transport infrastructure: Jonathan's plan for a complementary asset to pull heavy trucks off a congested free interstate and onto a toll road, built to stand on its own. The existing driver survey described the pain but couldn't price anything, so he designed and orchestrated a conjoint choice model and a pricing study, with a bias review alongside every draft of the questionnaire, because a badly worded question corrupts the signal it is meant to capture. Consumer product: the same method on a shelf product.
Marketing, measured. In the infrastructure case, the plan was Jonathan's: a travel plaza as a complementary asset to pull heavy trucks onto a toll road, built to stand on its own. The company's existing intercept survey described the pain but couldn't price a product, so before committing capital he set out to measure what would change a driver's route, who actually makes that decision, and what drivers would pay. Every combination of the product's attributes would have meant 864 profiles, so the design bundled what the plaza would offer anyway into a base product and pruned to 17. Each draft of the questionnaire carried a bias review: did “gas” mean fuel or fuel plus a store, did respondents picture one plaza or truck stops nationwide, was “clean” absolute or relative. The answers reframed the pricing: most drivers don't pay their own tolls, so a 20% toll cut was a weak lever that would have cost about 14% of the toll revenue, while fuel on site moved more drivers and kept it. Price ladders and elasticity set the amenity prices where revenue peaks — $18 parking and $16 showers against the $12 the business case had assumed. Jonathan orchestrated the whole of it; he and one analyst did the design, the questions, the analysis and the pricing. The consumer case runs the same method on a hypothetical product, adding what shelf products bring: correlated claims, perceived freshness, and the gap between the revenue peak and the profit peak. Both cases are presented as illustrative examples of the method. Elasticities are computed from the tested price points; every name, place, brand and firm is stripped; willingness figures are stated intent, not observed behavior.
Gaps filled. Problems solved. Work executed — and validated against the numbers. That is the engagement.
EXHIBITS A–Q ARE REPRESENTATIVE SCHEMATICS. STRUCTURAL COUNTS ARE TAKEN FROM THE ACTUAL ARTIFACTS; FIGURES ARE FROM THE WORK ITSELF OR MARKED ILLUSTRATIVE; EVERY NAME AND IDENTIFIER IS STRIPPED.