Firefly Aerospace

Phase A — Understand the business

Lens 1 · Company Overview

Firefly Aerospace is a vertically-integrated "responsive space" company selling end-to-end cislunar services across three product lines that ladder up in size: small launch (Alpha), lunar landers (Blue Ghost), and orbital transfer/servicing vehicles (Elytra), now bolted to a defense-software arm (SciTec). The pitch is to be the only Western "one-stop shop" that can launch a payload, move it on-orbit, and land it on the Moon — with a single vertically-integrated manufacturing base in Briggs, Texas.

• Alpha — a ~1,030 kg-to-LEO small launcher; pitched as "the only operational U.S. 1-ton launcher" and the first commercial rocket to launch on 24-hour notice. Reliability is the problem (Lens 3).
• Blue Ghost — the lunar lander. Blue Ghost Mission 1 became the first fully successful commercial soft-landing on the Moon (2 Mar 2025, Mare Crisium), delivering 10 NASA CLPS instruments. This is the crown jewel and the single most important credibility event in the company's history.
• Elytra — orbital transfer vehicles (Dawn/Dusk/Dark) for on-orbit mobility, payload hosting, lunar imaging ("Ocula", using LLNL telescopes) and "cislunar domain awareness".
• Eclipse (formerly MLV) — a medium-lift launcher (~16,300 kg to LEO) co-developed with Northrop Grumman, first flight targeted "as early as 2026".
• SciTec — acquired 2025 (~$855M); missile-warning / tracking / low-latency-AI software, the vehicle into the Golden Dome missile-defense program.

Customer & contract structure. Government-anchored: NASA (CLPS lunar delivery — Blue Ghost M1, M2, M4 at $176.7M, plus a $75M JPL "MoonFall" subcontract Jun 2026) and the U.S. Space Force / DoD (Eclipse → NSSL Lane 1; SciTec → Golden Dome Space-Based Interceptor OTA). Commercial: a Lockheed Martin multi-launch agreement, up to 25 Alpha missions over 5 years. Contracts are largely milestone/cost-plus government awards and fixed-price launch — backlog ~$1.4B at YE2025, up 23% YoY. Revenue recognition is therefore lumpy and program-milestone-driven (relevant to Lens 10).

Lens 2 · Supply Chain

Map: raw inputs → Firefly → end customer, named stakeholders. (web-only; no supply-chain.md exists in the research layer.)

Upstream inputs:

• Carbon composite — Firefly builds its own tanks/structures using an Automated Fiber Placement machine from Ingersoll Machine Tools (Camozzi Group) — claims all large Alpha composite structures in 7 days, MLV/Eclipse in 30 days. This is a genuine vertical-integration asset and a differentiator vs. peers who buy structures.
• Propulsion — in-house Reaver (Alpha 1st stage), Lightning (Alpha 2nd stage), and Miranda (Eclipse, >230,000 lbf class) engines; >60 Miranda hot-fire tests done. Firefly abandoned the aerospike to standardize on tap-off-cycle Reaver/Lightning. Historically partnered with Aerojet Rocketdyne.
• Avionics / structures / propulsion are vertically integrated at the Briggs, TX "Rocket Ranch" (expanded 92k → 207k sq ft for Eclipse).
• On-orbit compute — Elytra/Ocula uses NVIDIA Jetson edge processing; imaging telescopes from Lawrence Livermore National Lab.

Midstream (Firefly): integration in Briggs; launch from Vandenberg SLC-2W (Alpha) and Wallops/MARS, Virginia (Eclipse/Antares, with Northrop).

Downstream (end customers): NASA (CLPS/JPL), U.S. Space Force / DoD / "Department of War" (NSSL Lane 1, Golden Dome SBI), Lockheed Martin (Alpha rideshare/dedicated), Northrop Grumman (Eclipse co-development + Antares 330 for ISS Cygnus cargo).

Chokepoints / single-source dependencies: (1) Northrop Grumman is both a key supplier/co-developer AND a customer/investor for Eclipse — deep dependency on one partner for the medium-launch future. (2) Government revenue concentration — NASA + Space Force are the dominant buyers; a budget pivot (e.g., Artemis re-scoping) hits the top line directly. (3) Engine test infrastructure and skilled composite labor are internal bottlenecks to the 2x launch-cadence goal.

Lens 3 · Competitive Advantages (moats)

What's genuinely defensible:

• A landed-on-the-Moon track record. Blue Ghost M1 is the only fully successful commercial lunar landing to date — Intuitive Machines' IM-1/IM-2 tipped over. In CLPS that is a hard, demonstrated capability competitors can't claim. This is the strongest moat in the portfolio.
• Vertical integration + in-house composites/propulsion — owning AFP composite fabrication and its own engine family compresses cost and cadence and is hard to replicate quickly.
• Government entrenchment — CLPS incumbency, the Northrop Eclipse tie-up, and now SciTec's Golden Dome OTA create switching costs and program lock-in; defense primes prefer flight-heritage suppliers.
• The Northrop Grumman relationship — a prime validating, funding ($50M Series D into Eclipse), and buying the medium launcher is a moat and a dependency.

Where the moat is thin / bargaining power is weak:

• Alpha's reliability record is a reverse-moat: 4 failures in 6 flights since 2021. A launch vehicle that fails two-thirds of the time has negative brand equity with commercial customers and cedes pricing power.
• Against SpaceX (now public, gravitational pull on the whole sector) and a well-capitalized Rocket Lab (Neutron medium-lift due late-2026), Firefly is a price-taker, not a price-maker, in launch.
• Bargaining power over its anchor customers (NASA, Space Force) is low — they are near-monopsony buyers of cislunar/defense services.

Lens 4 · Segments

No segments.csv in the research layer — all figures /, not broken out cleanly by the company in public summaries.

Firefly does not publish a clean public revenue-by-segment table in the summaries available; revenue is reported on a consolidated basis with backlog color. Qualitatively the mix in 2025–26 is:

• Lunar (Blue Ghost/CLPS) — the largest and most visible driver in 2025; Blue Ghost M1 success plus M2/M4 awards drove the 163% FY25 revenue jump.
• Launch (Alpha) — small and impaired by the failure record; not a near-term growth engine.
• Spacecraft/Elytra + defense (SciTec) — the intended growth mix shift: SciTec (defense software) is now consolidated and the MoonFall/Golden Dome awards point the 2026 mix toward defense + lunar services, away from pure launch.

Trend & cause: revenue is accelerating hard off a tiny base (Q4-24 $9.0M → Q4-25 $57.7M, +541% YoY; FY24 $60.8M → FY25 $159.9M, +163%). The acceleration is driven by lunar-mission milestone recognition and the SciTec consolidation — not by a healthy launch cadence. That is the single most important thing to understand about the "growth" here.

---

Phase B — Measure performance

Lens 5 · Earnings Result (latest print — Q1 2026, reported 4 May 2026)

• Revenue $80.9M, +45% YoY (vs $55.9M Q1-25).
• Net loss $96.7M (vs $60.1M loss Q1-25) — the loss widened as revenue grew.
• GAAP gross margin 21.6%, DOWN from 27.7% in Q4-25 — margin went the wrong way QoQ.
• GAAP operating loss $95.7M (vs $85.6M Q4-25).
• EPS −$0.61.
• Liquidity — total liquidity $811.6M incl. $551.6M cash & equivalents; operating cash burn −$62.5M in the quarter. ⚠ Conflict flag: one cash-flow line in the same source set reads cash "$793.0M → $326.2M over the quarter" — likely a restricted-cash reclass or capital-deployment swing (SciTec close, capex); the $551.6M cash headline is used here, but the gross cash movement signals heavy deployment beyond the −$62.5M operating line. Verify against the 10-Q cash-flow statement on the next research-layer pass.
• Backlog ~$1.4B (YE2025).
• Guidance — FY2026 revenue $420–450M reaffirmed (~+170% at midpoint vs FY25 $159.9M).
• Market reaction: the stock has rallied on contract wins (Q1 print + Golden Dome news → +15%; MoonFall NASA award Jun 2026 → +~21%). The tape reacts to contracts and government catalysts, not to the loss line (Lens 8).

Unusual vs. own history: revenue is compounding fast but losses and cash burn are scaling with it — Q1-26 net loss $96.7M is the largest quarterly loss in the public record, and gross margin compressed. This is a pre-scale hardware/defense company funding a multi-year capability build (Eclipse, Blue Ghost cadence, SciTec integration) — not a business near breakeven.

Lens 6 · Earnings Calls (sentiment trend)

No transcripts/ in the research layer; sentiment inferred from public call coverage.

Management's narrative across Q2-25 (post-IPO) → Q1-26 is a deliberate re-anchoring from "small launch company" to "defense + cislunar services platform." Recurring themes that grew: Golden Dome / SciTec / national security; backlog quality and government awards; Eclipse first-flight readiness; "responsive space." Themes that receded: Alpha-centric messaging (understandably — the failure record), pure commercial-launch TAM. Tone is promotional and catalyst-forward — every print is paired with a contract headline (Golden Dome with Q1, MoonFall with the June offering). The Q3-25 call leaned on the per-share loss improvement ($3.57 → $1.50, "−58%") as the headline rather than the absolute burn — a tell that management is steering attention to rate-of-change, not level. Read sentiment as confident-but-pivoting: the leadership knows launch alone won't carry the multiple and is loudly repositioning toward defense.

Lens 7 · Comps

Multiples are / with date, or n/a. NEVER fabricated. Firefly is unprofitable → P/E and ROE are n/m; EV/Sales is the operative lens for the cohort.

Read: On trailing sales Firefly looks extreme (~32x FY25); on forward sales (~12x FY26 guide) it sits between LUNR (~6x, cheaper, bigger lunar revenue) and RKLB (~94x, far richer, more mature franchise). The cohort de-rated on the SpaceX IPO. The honest framing: FLY is priced as a defense-space growth story, not a launch company — at ~12x forward sales for a business burning ~$100M/quarter, the multiple needs the FY26 ~+170% growth and the Golden Dome/defense narrative to both deliver.

Lens 8 · Stock-Price Catalysts (moves that matter)

Firefly has only ~10 months of tape (IPO 7 Aug 2025), but the pattern is already clear — it is a headline/contract stock, not a fundamentals stock:

• 7 Aug 2025 — IPO. Priced $45 (above $41–43 range), raised ~$868M; opened ~$70, closed $60.35 (+34%), intraday valuation peaked ~$9.8B.
• Late 2025–Q1 2026 — the de-rate. Stock slid from ~$60 toward the high-teens; 52-week low $16.00. Market cap fell ~−43% IPO→Mar-2026.
• SpaceX IPO day (~Jun 2026) — −19% in a day on a sector-wide space selloff; KeyBanc called it a buying opportunity.
• Contract wins = the up-catalysts: Golden Dome/SciTec OTA + Q1 beat → +15%; MoonFall $75M NASA award (Jun 2026) → +~21% and a KeyBanc upgrade to Overweight.
• Dilution = the down-catalysts: the $48.00 secondary (12M shares: 4M primary + 8M selling holders, closing ~1 Jun 2026) and the SciTec resale registration (~11.1M shares) both pressured the stock on overhang fears.

What the market actually reacts to: government contract awards (up) and share-supply/dilution events (down). Alpha launch outcomes matter for credibility but the stock has largely looked past the failures toward the defense/lunar story.

---

Phase C — Judge people & books

Lens 9 · Management

• CEO — Jason Kim (since Oct 2024). 25 years in space/defense; ex-CEO of Millennium Space Systems (Boeing's smallsat unit, which he scaled), prior senior roles at Raytheon Intelligence & Space and Northrop Grumman Aerospace Systems; USAF Academy (BS EE), USAF officer 1999–2006, MS EE, UCLA Anderson MBA. Read: a defense-industrial operator, not a founder-engineer — and his pedigree explains the whole strategy: the pivot to Golden Dome/SciTec and the Northrop tie-up are exactly the playbook of someone who came up through the primes. Right archetype for a government-anchored space company at the scaling stage.
• COO — Ramon Sanchez (from Dec 2025). 12+ years at Boeing, latterly senior director of operations for space, intelligence & weapon systems — hired explicitly to drive production scaling and execution. Signals management knows the bottleneck is manufacturing throughput.
• Tenure & skin in the game: leadership is new (CEO <2 yrs, COO ~6 months). Insider economics are dominated by the sponsor, not management — see ownership.
• Capital allocation: the defining move is the ~$855M SciTec acquisition — funded largely with IPO cash + stock (hence the resale overhang). Bulls call it a "steal" for a Golden Dome foothold; bears see a large, partly-stock-funded deal that deepened dilution and pivoted the story barely a year after listing. ROE/ROIC are negative — there is no value-return track record yet; this is a deploy-everything-into-capability phase.
• Ownership / control — the key governance fact: AE Industrial Partners (the PE sponsor that took majority control in 2023) holds ~36.7–41.8% post-IPO and 5 of 9 board seats. Firefly is a sponsor-controlled company. Minority public holders have limited governance power, and the selling-stockholder secondaries are partly the sponsor monetizing — a structural overhang for years.
• Red flags (governance): controlled-company structure; rapid post-IPO secondary + acquisition-driven dilution; promotional, catalyst-timed communications. None are accounting fraud signals — they are alignment/overhang risks.

Lens 10 · Forensic Red Flags

Forensic lens. web-only — no filings on disk; figures ``. Re-run against the 10-K/10-Qs once ingested (CIK 0001860160).

• Revenue recognition / quality: revenue is milestone- and program-driven (CLPS landers, cost-plus/OTA defense). Lumpy recognition + a fresh acquisition (SciTec) make YoY comparisons noisy and create room for timing-driven beats. Watch for revenue outrunning cash collection (unbilled receivables/contract assets).
• Margin volatility: GAAP gross margin swung 27.7% (Q4-25) → 21.6% (Q1-26) — thin and unstable, consistent with early-stage hardware + mission-cost absorption. Not a flag per se, but it means "gross margin" is not yet a reliable signal.
• Cash vs. earnings divergence: operating burn −$62.5M/qtr against a $96.7M net loss; the flagged gross cash movement ($793M→$326.2M) vs. the $551.6M cash headline needs reconciliation against the cash-flow statement — possible restricted-cash/escrow (SciTec) or heavy capex. Highest-priority forensic item to verify.
• Goodwill / intangibles: the $855M SciTec deal will sit largely as goodwill + acquired intangibles on a company with a ~$5.2B cap — future impairment risk if the Golden Dome program slips or SciTec underperforms.
• Stock-based comp & dilution: post-IPO SBC plus secondary issuance (4M primary) plus an 11.1M-share SciTec resale registration → persistent share-count growth; non-GAAP metrics (if used) will be flattered by SBC add-backs. Track diluted share count quarterly.
• Going-concern: not an acute risk — ~$551M cash + $260M revolver and capital-markets access (IPO + secondary) give multi-quarter runway at the current burn. But at ~$100M/qtr losses, further raises are a when-not-if, and each is dilutive.

Regulatory findings (required sub-section).

• SEC enforcement (EDGAR LR/AAER): None. Per the Stage-1 file regulatory/regulatory-findings.md (fetched 2026-06-17): Firefly had no CIK at fetch time (pre-IPO record) so no EFTS search ran; total SEC findings = 0. Post-IPO, FLY now files under CIK 0001860160 — a clean EDGAR enforcement search should be run on the next pass.
• Non-SEC enforcement (web search "Firefly Aerospace" (FTC OR DOJ OR FDA OR CFPB OR consent decree OR settlement OR fine OR penalty) enforcement): No material enforcement actions, consent decrees, or fines surfaced. The only "investigations" are routine FAA mishap investigations of Alpha launch failures (e.g., Flight 6 / FLTA006; FAA cleared return-to-flight 26 Aug 2025) — these are standard launch-license safety reviews, not enforcement penalties. (Note: an earlier, different Firefly entity faced a 2010s ITAR/export ownership issue that led to its bankruptcy and re-formation under Noosphere/Polyakov, later resolved when AE Industrial took control — historical, not current; verify if it matters to a thesis.)
• 10-K Item 3 (Legal Proceedings): Not available — no 10-K on disk; must be read from the filed annual report on the next ingest.
• Summary: No material regulatory/legal findings surfaced via the Stage-1 SEC file (0 findings, pre-CIK) + web search as of 2026-06-17. Caveat: this is web-only and pre-filing-ingest — re-verify Item 3 and run a post-IPO EDGAR LR/AAER search once FLY's filings are on disk.

---

Phase D — Project & stress-test

Lens 11 · Forward Projection

Bottom-up from FY25 actual + FY26 guide. All inputs ; outputs with arithmetic. Company is loss-making → projection is revenue + path-to-profitability, not EPS-positive. No Brier forecast logged (breadth/watchlist loop).

Anchors: FY25 revenue $159.9M (+163%); FY26 guide $420–450M (mid ~$435M, ~+170%); backlog $1.4B; Q1-26 burn ~$96.7M net loss; ~164M shares.

• FY2026 (base): revenue ~$435M (company mid). Net loss ~$330–380M. EPS ~−$2.0 to −$2.3.
• FY2027 (base): revenue ~$650–750M. Loss narrows to ~−$200–260M as gross margin climbs toward ~30%+ on volume. EPS ~−$1.1 to −$1.5.
• FY2028 (base): revenue ~$950M–1.1B; approaching gross-profit-positive at scale, still operating-loss or near-breakeven. EPS ~−$0.4 to +$0.2.

Bull path: Eclipse flies clean in 2026 and wins NSSL Lane 1 task orders; Golden Dome SBI converts from OTA to a funded production program; Blue Ghost cadence holds → FY28 revenue >$1.3B, operating breakeven pulled into 2027. Bear path: Eclipse first flight slips or fails (Alpha's record makes this the base-rate worry); Golden Dome budget gets cut/re-competed; Alpha stays grounded/unreliable → FY26–27 revenue misses the ~+170% guide, burn forces another dilutive raise, multiple compresses toward LUNR's ~6x.

The number that actually matters (runway-to-catalyst, +private discipline retained): with ~$551M cash + $260M revolver against ~$100M/qtr burn, runway is ~6–8 quarters absent a raise — i.e., cash reaches the Eclipse first-flight and the Golden Dome 2028 demo decision windows, but only just, and any slip likely triggers a dilutive raise first.

Lens 12 · Bull vs Bear

Bull case. Firefly is the only commercial team that has actually landed on the Moon, and it's parlaying that credibility into the two richest government tailwinds in space: NASA CLPS/Artemis lunar logistics and the Golden Dome missile-defense build-out. Vertical integration (in-house composites + engines) gives it cost/cadence control few peers have. The Northrop Grumman partnership de-risks Eclipse (funding + a guaranteed Antares/Cygnus use-case + prime credibility), and SciTec is a cheap option on a multi-hundred-billion-dollar defense program ($855M for a Golden Dome SBI seat among only 12 awardees). At ~12x forward sales for ~+170% growth, the price is defensible if you believe the defense pivot. Earnings surprises come from contract awards, and the award cadence (MoonFall, Golden Dome OTA) is accelerating.

Bear case (permanent-impairment risks). (1) Alpha can't be trusted — 4 failures in 6 flights. A launch company whose rocket fails two-thirds of the time has no commercial-launch moat, and the same engineering culture is building Eclipse. (2) It's a money furnace — ~$100M/qtr losses, margin going backwards QoQ, and a structural dependence on serial dilutive raises; the $48 secondary + 11.1M SciTec resale shares are a multi-year overhang. (3) The whole thesis now leans on Golden Dome — a political program (budget-and-administration-dependent) that may be cut, delayed, or re-competed, with SciTec sitting on the balance sheet as ~$855M of impairable goodwill. Pre-mortem (18 months out, thesis broke): Eclipse's first flight slipped to 2027 or failed; Golden Dome funding got trimmed in a budget fight; FLY missed the FY26 revenue guide; a forced raise at a low price gutted the multiple; the stock sits near its $16 low.

Are multiples too high? On trailing sales, yes (~32x). On forward sales (~12x) it's reasonable only if FY26 ~+170% growth lands and the defense narrative holds — i.e., the price embeds near-flawless execution from a company with a flawed launch record. Contrarian view — what the market refuses to see: the bull/bear debate is fixated on rockets, but Firefly has quietly become a defense-software-and-missile-defense play wearing a rocket company's clothes. SciTec + Golden Dome may matter more to the 2027–28 P&L than Alpha or even Eclipse — and that's both the upside (re-rate as a defense-tech name) and the risk (single-program political dependence).

Lens 13 · Devil's Advocate (short-seller)

Dismantling the bull case.

• Structural break in how it makes money: Firefly doesn't have a profitable way to make money yet — it converts government milestones and an acquired software book into revenue while burning ~$100M/qtr. Remove capital-markets access (a space-sector risk-off, à la the SpaceX-IPO selloff) and the model doesn't self-fund.
• Revenue concentration: overwhelmingly government (NASA + Space Force) + two primes (Lockheed, Northrop). If Artemis is re-scoped or Golden Dome is trimmed, a huge share of backlog/pipeline is exposed at once. This is monopsony risk, not diversification.
• The moat is weaker than bulls think: the lunar moat is real but narrow (a handful of CLPS missions a year); the launch moat is negative (failure record); the defense moat is borrowed (SciTec's, and dependent on a political program).
• Most dangerous competitor bulls underestimate: Rocket Lab. Neutron (medium-lift, reusable, late-2026) attacks Eclipse's exact niche from a far stronger balance sheet and a reliable small-launch heritage (Electron). And SpaceX, now public, will price-compress the entire launch market.
• Worst capital-allocation move: an $855M, partly-stock-funded acquisition of a software company barely a year after IPO, pivoting the story and deepening dilution — with AE Industrial (5/9 board) selling into secondaries alongside.
• Assumptions that must hold for today's price: Eclipse flies (on time, successfully) in 2026; FY26 revenue hits ~$435M; Golden Dome stays funded; no further large dilutive raise near current levels.
• If growth disappoints 20–30%: FY26 revenue ~$300–350M instead of $435M → the forward multiple jumps back toward ~15–17x on a decelerating loss-maker → de-rate toward the $16–20 range, probably accompanied by a forced raise.
• Single scenario that permanently impairs: Eclipse's first flight fails (Alpha base-rate) and Golden Dome funding is cut in the same 12 months — credibility and the defense thesis break together, goodwill impairs, and the equity re-rates to a small-cap launch also-ran.

Lens 14 · Management Questions (ordered by information value)

1. Eclipse first flight: what is the honest probability it flies before end-2026, and given Alpha's 4-of-6 failure record, what specifically in design/test/culture makes you confident the first Eclipse flight succeeds?
2. Golden Dome / SciTec: what is the realistic funded-program path from the OTA to a production contract, and what does your 2027–28 revenue look like if Golden Dome funding is cut by 50%?
3. Cash bridge: reconcile the Q1-26 cash movement ($793M→$326.2M) with the $551.6M cash headline and the −$62.5M operating burn — where did the cash go, and when is the next raise?
4. At ~$100M/quarter losses, what is the explicit path and date to operating breakeven, and what gross-margin level does it require?
5. Alpha: is small launch still a strategic priority or effectively a sunk-cost capability — and would you discontinue it if Eclipse succeeds?
6. SciTec: what were the actual 2025 revenue/margins of the acquired business, and how much of the $1.4B backlog is SciTec vs. Firefly-organic?
7. AE Industrial controls 5 of 9 board seats and is selling into secondaries — how do you protect minority holders, and what's the sponsor's exit timeline?
8. Northrop Grumman is supplier, co-developer, customer, and investor on Eclipse — what happens to Eclipse economics if that relationship changes?
9. NSSL Lane 1: how many task orders do you realistically expect to win, against what competition (Rocket Lab Neutron, others), and at what margin?
10. Revenue recognition: how much current revenue is milestone/percentage-of-completion vs. delivered, and how should we think about quarterly lumpiness?
11. Goodwill from SciTec: what are the impairment triggers, and at what point do you re-test?
12. Lunar cadence: Blue Ghost is targeting 2x missions — what is the per-mission gross margin trajectory as you move down the cost curve?
13. SpaceX is now public and will compress launch pricing — where do you have durable pricing power, if anywhere?
14. Dilution: with the secondary + the 11.1M SciTec resale shares, what's your expected fully-diluted share count at end-2027?
15. If you had to pick the single capability that justifies a >$5B valuation three years out, is it launch, lunar, or defense — and why?

---