Sustainable Finance — Silver Leaf Partners

Summer 2025 on the investment team at Silver Leaf Partners, a New York–based firm focused on the companies and infrastructure of the climate transition. Twelve weeks, three workstreams: financial modeling, deal-pipeline research, and a CRM rebuild. Reporting line was directly to the Managing Partner.

Modeling

The bulk of my time. DCF and project-finance work on early-stage targets across three sub-strategies the firm tracks — sustainable infrastructure, carbon removal credits, and energy-transition adjacencies.

• DCF builds for early-stage sustainable-infrastructure companies — revenue forecasts under base / upside / downside cases, three-statement integration where the cap structure mattered, sensitivity tables that surfaced the two or three assumptions actually doing the work
• Tax-equity and project-finance sensitivities on renewable-energy deal structures the team was diligencing
• Supported the IC Fund_I pitch deck: my job was to translate thousand-row models into slide-level positions — without losing the honest middle of the range. The deck went through six rounds of revision; what survived was the version that said the least and meant the most

The trick of the modeling work, the one I had to learn the slow way, is that a fund pitch is not a math problem. The math has to be defensible, but what you are actually selling is judgment under uncertainty. The slides exist to demonstrate the team's reasoning, not the model's precision. Models that try to argue from precision get caught — IC committees know better. Models that argue from a narrow, defensible range, with the assumptions named, are the ones that land.

Pipeline research

Every week I read the deal flow — private-markets databases, climate-fund newsletters, the press releases from operators we tracked — and wrote short briefs on what mattered for the fund's thesis.

The categories I came back to most:

• Carbon removal: Frontier's offtake commitments and the deals running through Arbor and Climeworks; Rubicon Carbon's secondary purchases and the resale market that's quietly forming around them; Deep Sky's Canadian DAC build-out and what Quebec's hydro economics imply about the unit cost
• Energy-transition credits: Kinetic Coalition's early coal retirement deals — interesting because they monetize avoided emissions through a structure that didn't exist three years ago, and because the market hasn't decided yet whether that's a real asset class or a clever press release
• Adjacencies: grid software, battery-as-a-service deployments, the small EPCs consolidating in the IRA tailwind. Most of these were file-and-watch, not file-and-pitch
• PitchBook research on manager performance and social-impact mandates — useful for the fund's LP outreach, where matching mandate to thesis is half the battle
• Market-size work for two of the firm's target segments

Most of the briefs were short and most went unused. The point wasn't every brief; the point was that the partner, on a Tuesday morning, could ask a question about a sub-segment and get an answer by Wednesday.

CRM rebuild

Firms with small teams feel their own systems. What gets logged, gets remembered. When I arrived the Client Prospect Investors workbook was a single Excel sheet with thousands of rows, no version control, and a column called "notes" that contained the entire history of investor relationships in free text.

My standing mandate for the summer was to rebuild it. The new workbook:

• Maps every conversation to a fundraising stage (cold / warm / pitching / diligence / committed / closed)
• Tags investor profiles by check size, mandate fit, geography, and decision-cadence
• Wires the deal archive back to the contact records, so a follow-up email two quarters later lands with context — we last spoke in May, you mentioned you were waiting on a board meeting, the deal you flagged closed in August at $34M
• Keeps the free-text notes column, because some things genuinely don't fit a schema

I left it better than I found it. The partner used it on the day I left and the week after I left, which is the only test that matters for an internal system.

What it taught me about pipelines

A pipeline is a memory system. The honest version of that deal didn't go anywhere is almost always we lost the thread. Most deals don't die from a "no" — they die from drift. The CRM is what stops the drift, and the CRM works only if the team uses it on a Tuesday when nothing is on fire.

Confidentiality

Specifics about diligence calls, target companies under active LOI, and LP names are deliberately not on this page. The thinking is shareable; the deals are not.