$AMPG: Forty-Eight Percent Gross Margins, Zero Debt, And A Five Dollar Print Nobody Trusts Yet
AmpliTech is a twenty-four-year-old RF house that turned itself into a 5G radio OEM with a quantum tail. Q1 2026 revenue plus forty-eight percent, gross margin up fifteen hundred
Shawarma Capital. May 25, 2026. AMPG is $5.04 against a Friday close. I have no AMPG position as of this writing. I am opening a starter position via daily buys Tuesday May 26 through Friday May 29, 2026, sized off stable holdings, with the conditional add line set below. This post is research synthesis, not investment advice.
Almost every microcap that gets pitched as a 5G play is a parts house that drafts off Ericsson, Nokia, or Samsung. They sell components. They do not sell radios. When a hyperscaler or a Tier 1 carrier writes the check, the parts house gets a single line on a bill of materials and a margin the buyer pre-negotiated.
AmpliTech Group is not that company anymore. It used to be. It is not now.
Q1 2026 closed with $5.35 million of revenue against $3.60 million a year earlier, a 48.6 percent year-over-year print. Gross profit doubled. Gross margin moved from 33.0 percent to 48.0 percent in a single year, a fifteen-hundred-basis-point step-up that the tape has not yet priced. Cash plus marketable securities sits at $18.4 million. Working capital is $25.4 million against $10.2 million on December 31. Stockholders' equity is $48.4 million. Long-term debt is zero. The company funded its 2026 build through a $16 million combined rights offering and registered direct, both closed before this quarter ended. The dilution has already happened.
AmpliTech sells more than parts now. It sells a 64T/64R massive MIMO ORAN 5G radio unit, the only fully integrated open-RAN small-cell hardware that a US-domiciled vendor manufactures end-to-end on US soil. It sells cryogenic low-noise amplifiers, the kind every qubit on every superconducting quantum computer in the IBM, Rigetti, and IonQ pipelines requires. It sells GaN power amplifiers and ultra-low-noise satcom RF for Viasat, Lockheed Martin, and L3Harris. And it sells managed Communications-as-a-Service through a subsidiary, True G Speed Services, that exists to convert the radio hardware into a recurring revenue stream a Tier 1 carrier can actually buy.
The market cap at $5.04 against 25.34 million shares outstanding is roughly $127.7 million. Enterprise value, after backing out cash, sits near $109 million. Trailing twelve-month revenue is somewhere between $14 million and $16 million depending on how you handle the 2025 reset quarter, with Q1 2026 running at a $21.4 million annualized pace. Management has not retracted full-year 2026 guidance. CEO Fawad Maqbool said on the May 13 call that the guidance remains achievable, with the revenue ramp weighted to the second half. He owns 2,663,864 common shares directly. He owns more than ten percent of the company he founded. He has never sold a share since the IPO. He bought into the rights offering this January at his pro-rata allocation.
I am not telling you Maqbool has a clean execution record. He does not. The 2024 fiscal year was a hiring miss against a $30 million guide that landed near $10 million. The company lost $3.5 million in a custodial Bitcoin treasury arrangement to fraud the FBI is still investigating. The bear case from the original short writeups, the ones that traded the stock down to $2.75 in late 2024, is grounded in CEO credibility problems that are real and recent. I will not paper over them in this piece. They are in Section X.
What changed between $2.75 and $5.04 is the unit economics. Forty-eight percent gross margin is what a 5G radio OEM looks like at scale. It is not what a custom-microwave parts house looks like, ever. The mix shift has happened. The print proves it.
Let's go.
I. The Print
The Q1 2026 release dated May 13, 2026 disclosed four numbers that matter and three the tape did not parse correctly.
The four that matter:
The three the tape did not parse:
The manufacturing and engineering segment, the operating core that produces revenue, generated $3.28 million of Q1 revenue against $0.99 million in Q1 2025. That is a 231 percent year-over-year acceleration. The remaining $2.07 million ran through the True G Speed Services subsidiary and adjacent services lines. The mix is real. The pivot to integrated 5G radios shows up cleanest in the segment breakdown, not in the headline.
Working capital rose from $10.2 million on December 31, 2025 to $25.4 million on March 31, 2026. The current ratio went from 1.68 to 4.25. That is not a working-capital ramp from operations. It is the $16 million of fresh equity the company raised through a combined rights offering plus registered direct, both pre-funded before the quarter ended. Cash and marketable securities ended the quarter at $18.4 million. Long-term debt remained zero, where it has been for the entire post-IPO history. The balance sheet is the cleanest in the company's filing history.
Full-year revenue guidance is intact but weighted to the second half. Maqbool said on the May 13 call: "The Company continues to believe its full-year revenue guidance remains achievable; however, the Company expects revenue recognition to be more heavily weighted toward the second half of the year." There is no retracted number. There is no missed guide. There is a calendar slip.
The structural read: a parts-house revenue base at 33 percent gross margins is not the same business as an integrated-systems revenue base at 48 percent gross margins. The Q1 print is the first quarter where the systems-revenue mix carried the margin line. The rest of the year is the test of whether that mix keeps.
II. The Company
AmpliTech Group was founded in 2002. Twenty-four years old. Headquartered in Hauppauge, New York, on Long Island. Public on Nasdaq since 2021. Approximately one hundred fifty employees across engineering, manufacturing, and sales.
The founder is Fawad Maqbool. He is CEO, chairman, and beneficial owner of more than ten percent of the company. His background is the back-half of the New York RF and microwave industry: Hazeltine Corporation from 1983 to 1986 (acquired by Emerson, now BAE Systems), Miteq from 1987 to 1999 where he ran a department of thirty-two engineers (acquired by L3Harris in 2015), and AmpliComm from 2000 to 2001, a fiber-optic and cryogenic amplifier shop he founded that Aeroflex acquired within a year. He founded AmpliTech in 2002 with an explicit RF and microwave focus. He earned two bachelor's degrees from CUNY and a master's in electrical engineering from NYU. He has never sold a share of AMPG since the IPO. He has bought multiple times on the open market. He voluntarily cut his salary 20 percent in 2024.
The company organizes around four product lines:
The Hauppauge facility manufactures the components, integrates the radios, and ships the systems. The vertical integration claim Maqbool makes on every earnings call is that no other US-domiciled vendor produces ORAN 5G radios end-to-end on US soil. That is a Buy America claim, and it is the most important fact in the entire setup because the BEAD program rolling out through 2027 requires 65 percent US-sourced components today, rising to 75 percent by 2029. A vertically integrated US OEM does not have to assemble a coalition of foreign suppliers to satisfy the requirement. It already does.
The four-line product strategy is not new. The integrated systems revenue mix is.
III. The Pivot
For twenty years, AmpliTech sold parts. It built ultra-low-noise amplifiers for satellite ground stations, custom MMIC packaging for defense customers, and one-off design wins on radar and electronic-warfare programs. The revenue base was lumpy, capped by the size of the largest single contract, and structurally limited to component-level margins. Net income for the decade preceding the IPO oscillated around zero.
The pivot started after the 2021 IPO. Two events drove it. First, the Open RAN telecom standard reached commercial deployment readiness in 2023 and 2024, anchored by the O-RAN Alliance specifications that decoupled the radio unit, the distributed unit, and the centralized unit into separately specifiable hardware. That decoupling created room for a non-Ericsson, non-Nokia, non-Samsung vendor to sell an ORU specifically. AmpliTech already manufactured the RF front-end that goes into an ORU. The leap from front-end to full ORU was an integration step, not a clean-sheet design problem.
Second, the rollout of the federal BEAD program, the $42.45 billion broadband expansion authorized under the Infrastructure Investment and Jobs Act, codified domestic-content requirements that effectively excluded the European and Asian incumbents from the rural-broadband buildout. A US-domiciled radio vendor with a finished ORU and a managed-services wrapper was the exact category the BEAD funds were structured to favor.
The True G Speed Services subsidiary, branded TGSS, exists to package the radio into something a Tier 1 telco can buy as a service. Press releases from 2024 describe it as managed Communications-as-a-Service for ORAN networks and base stations. The model is the same model a hyperscaler runs: AmpliTech owns the hardware, the customer pays a recurring fee per radio per year for managed operations. The recurring revenue line is what converts the historical lumpy parts business into the kind of cash-flow profile a sell-side analyst can model.
The 64T/64R massive MIMO ORAN radio is the flagship product. The published throughput is one gigabit per second. The form factor and the power envelope target small-cell deployments, the segment of the 5G network that the Tier 1 carriers consistently undershoot because the incumbent vendor stacks were not designed for granular site rollout. The radio is the SKU that lets a telco bid a BEAD project, drop one box per site, and meet domestic-content rules out of the gate.
The pivot from parts house to integrated radio OEM is the thesis. The Q1 print is the first quarter where the integrated-systems mix actually moved the margin line. Three more quarters at 48 percent gross margin and the rerate is no longer optional. It is the first thing a generalist tech analyst initiates coverage on.
IV. The Cap Table And The Tape
Shares outstanding sit at 25,340,000 as of the most recent disclosure. Float is approximately 21.0 million, derived from the 5.82 percent short-of-float ratio against the 1,222,778 shares reported short on the April 30, 2026 settlement date. Insider ownership totals roughly 30 percent across Maqbool plus the board and named officers. Maqbool's direct stake is 2.66 million shares, 10.5 percent of outstanding.
Two things stand out on that table.
The first is the average daily volume against the float. 3.45 million shares trade per day against a 21 million share float. That is 16.4 percent of the float turning over every session, which is unusually high. The name is liquid for its market cap. A position can be built and exited inside a single day's volume. That cuts both ways: every news event prints fast in either direction. The DCA structure I am using this week is built around that liquidity profile. I am not chasing.
The second is the short interest. 5.82 percent of float is a modest short. 3.1 days to cover is below the four-day threshold that historically anchors the borrow market. The shorts here are not running a thesis trade. They are running a hedge or a pair against the broader ORAN cohort. There is no squeeze setup, and I am not pitching one. The float math says you make money on this name through fundamentals, not through a borrow event.
The capital structure is clean. There is no convertible preferred. There is no ratchet provision sitting on a PIPE. The most recent capital raise was the $16 million combined rights offering plus registered direct that closed before the Q1 print. The rights offering was structured pro-rata to existing holders, with Maqbool participating to maintain his percentage. That structure is the cheapest possible dilution for an existing shareholder. It is what a CEO who knows he is going to own the company in five years agrees to.
There is an at-the-market offering facility still outstanding. Management has historically used it for housekeeping, not for material capital raises. The ATM exists. It can be tapped. Anyone reading this should track the post-effective amendment filings on EDGAR.
V. The 5G ORAN Lane
The Open Radio Access Network standard is the single largest commercial inflection in the telecom infrastructure cycle since long-term evolution.
The O-RAN Alliance, founded in 2018, published specifications that decompose a base station into separately purchaseable hardware. A radio unit, a distributed unit, and a centralized unit. Each can be sourced from a different vendor. The integration runs over open interfaces. The economics are the economics of any open standard: incumbent margins compress, new entrants get pricing room they did not have under the closed stacks.
The total addressable market for ORAN small-cell radios from 2026 through 2030 is somewhere between $30 billion and $50 billion depending on which research house you trust, with North American deployment running ahead of the global average because the FCC C-band and CBRS auctions front-loaded the spectrum allocation.
AmpliTech's flagship 64T/64R massive MIMO ORU is the SKU that competes in the 5G mid-band small-cell slot. Throughput targets one gigabit per second. Power envelope and physical footprint target the dense-urban and rural-coverage deployments where BEAD funding is flowing. The vendor list for that slot is short: AmpliTech, a handful of Asian ORU vendors who do not qualify for Buy America preferences, and the incumbent Ericsson and Nokia stacks that are not architecturally optimized for granular ORU sales. The Tier 1 carriers in the US run formal ORAN procurement programs at AT&T, Verizon, T-Mobile, and the regional incumbents. The procurement cycles run on twelve-to-eighteen-month qualification timelines. AmpliTech is past the qualification phase with the two named LOI customers from the 2024 cycle.
The 2024 LOI bench was disclosed at $118 million across two Tier 1 customers. The named one is Telus, the Canadian carrier, at $40 million. The unnamed second customer signed at $78 million. As of the Q1 2026 print, the company had recognized $12 million of firm purchase orders against the combined LOI base. The conversion math runs slowly because the qualification gates fire one site or one regional cluster at a time. The bear point that the LOI conversion has been slower than the 2023 framing implied is correct. The bull point that the conversion is moving, that gross margins on the converted revenue have stepped up to 48 percent, and that the second-half loading of the 2026 guide reflects the next named tranche, is also correct. Both can be true.
The BEAD program funding flow is the second mechanic. The $42.45 billion is being allocated by state Broadband Equity, Access, and Deployment Grant offices over the 2025 to 2028 window. The first major procurement awards landed in late 2025 and continue through 2026. The 65 percent domestic content rule applied to every radio purchased under BEAD funds, rising to 75 percent in 2029, creates a structural moat that no foreign-headquartered vendor can satisfy without restructuring its supply chain. AmpliTech's Hauppauge manufacturing site is one of the few US-domiciled facilities that integrates an ORU end-to-end. The Buy America claim is not a marketing line. It is a procurement-rule moat.
VI. The Quantum Lane
Every superconducting qubit on every quantum computer in the IBM Quantum, Rigetti, and IonQ roadmaps requires a cryogenic low-noise amplifier to read out its state without destroying the coherence. The amplifier sits inside the dilution refrigerator stack, operates at temperatures below one Kelvin, and has to deliver a signal-to-noise floor low enough not to swamp the qubit's quantum signal.
AmpliTech is the only US-based commercial vendor of cryogenic LNAs at production volumes. The per-unit pricing is between $5,000 and $7,000 depending on the band coverage and the noise figure. Maqbool has described the customer relationship publicly: every qubit on the IBM roadmap requires an LNA, and IBM has published a target of 100,000 qubits by 2033. The unit-economics math, even at the bottom of the price range and the optimistic end of the qubit count, is in the hundreds of millions of dollars of cumulative cryo-LNA revenue over the 2026 to 2033 window, before any contribution from Rigetti, IonQ, or the academic and national-lab market.
I will not put a base case on the quantum line. The IBM roadmap could slip. The qubit-count target is a 2033 number, not a 2027 number. The competing read-out architectures, the parametric amplifiers and the traveling-wave amplifiers that European and Japanese labs are developing, could displace cryo LNAs at the high-end qubit configurations.
What the quantum line provides is optionality. It is a multi-year structural tailwind on a product AmpliTech already manufactures, against a customer set that has no obvious alternative US supplier. Even modest penetration of the IBM target produces a revenue line that exceeds 2025 total company revenue. The quantum line is not the thesis. It is the tail.
VII. The Defense And Satcom Lane
Viasat is the largest single named customer in the historical filings. The relationship covers satcom ground-station LNAs and custom RF integration. The revenue contribution is approximately $5 million annually, structured as a multi-year framework against site rollouts on the Viasat ground station network.
Lockheed Martin and L3Harris consume custom microwave components for military satcom and radar systems. The revenue is project-driven. The 10-K language is generic: "custom microwave components for military satellite communications and radar systems." Concentration metrics on these customers are not disclosed at the contract level.
The defense and satcom lane is the historical base business. It is profitable on the component-level economics that the parts-house era ran on. It is not where the growth comes from. It is the floor under the quarterly revenue line that keeps the gross margin print stable while the ORAN and quantum lanes ramp.
VIII. The Customer Bench
The customer bench, named where disclosed:
That is the named bench. The unnamed customer pool, particularly in the cryo-LNA line where every superconducting quantum hardware vendor is a potential buyer, is wider than the disclosed list. Confidentiality agreements limit what AmpliTech can name in public filings. The Tier 1 unnamed ORAN customer is the most-asked question on every earnings call. Management has declined to name them because the carrier procurement cycle does not allow vendor announcements until the BEAD or commercial deployment is publicly committed. Track the FCC and state BEAD office award announcements through 2026 for the disclosure.
IX. The Competitor Map
The competitor map is different for each lane.
For the 5G ORAN small-cell ORU lane, the named competitors are Anokiwave (private, focused on phased-array antennas, recently acquired by Qorvo), Mavenir (private, software-led ORAN stack with hardware partners), Parallel Wireless (private, end-to-end ORAN), and the Asian ORU vendors who do not qualify for Buy America. The incumbent Ericsson and Nokia stacks compete only at the high-end macro-cell tier, not the small-cell ORU slot AmpliTech targets. AmpliTech's differentiation is the vertical integration plus the US manufacturing, not the radio specification per se.
For the cryogenic LNA lane, the competitors are Low Noise Factory (Swedish, the historical incumbent in academic and national-lab cryo LNA supply), Caltech-Microsemi (research-grade, not at production volume), and a handful of in-house programs at the major quantum hardware vendors. AmpliTech's differentiation is the US-based production volume and the commercial pricing point that scales with qubit count. The Swedish incumbent does not scale to the hundred-thousand-qubit target. AmpliTech can.
For the defense and satcom RF lane, the competitors are Qorvo, MACOM, Analog Devices Wireless, MaxLinear, Skyworks, and the captive RF programs at the defense primes. The lane is crowded. AmpliTech wins on small-program flexibility and on legacy customer relationships, not on volume economics. The lane is a margin floor, not a growth engine.
The cleanest comp for valuation purposes is the public small-cap RF and microwave cohort: Qorvo and Skyworks are large-cap reference points, MaxLinear and MACOM are mid-cap reference points, and Anokiwave (pre-acquisition) priced at a strategic multiple to Qorvo. The public 5G ORAN pure-plays are nearly non-existent in the small-cap segment: Mavenir is private, Parallel Wireless is private, and the listed ORAN names are component-level rather than full-stack. AmpliTech is the only US-listed integrated ORAN ORU vendor at a small-cap valuation. The comp set is more "compound semi growth" than "ORAN pure-play," because there is no clean ORAN pure-play in the small-cap public universe.
The named competitor map, simplified:
X. The Bear Case
I owe the bear case more space than the bull case in any Part 1. Here it is.
Maqbool has a credibility problem. The 2024 fiscal year was a $30 million revenue guide that landed at approximately $10 million. The miss was attributed to delayed FCC and CE certifications on the ORU product, which is a real and recurring risk in any radio-hardware ramp. The miss happened anyway. Anyone who entered the position in 2024 based on the $30 million number got a $2.75 print twelve months later. The 2024 guide miss is the single load-bearing concern on this name.
Layered on top of that, the company lost $3.5 million in 2024 to a custodial Bitcoin treasury arrangement that the FBI is investigating as fraud. The custodian failed. AmpliTech was a victim, not a perpetrator. But the decision to place corporate cash into a digital-asset custodian without sufficient operational due diligence is a capital-allocation judgment that any allocator has to weigh against everything else. It is a small dollar amount relative to the current cash balance. It is a large signal about the management team's risk-assessment process at the time.
The 2025 revenue guide of $24 million was implied rather than formally issued, and the year almost certainly came in below that number, with the bulk of the shortfall sitting in the second customer's order delays. The published Q1 2026 number annualizes to $21.4 million, with management explicitly guiding to a second-half load that needs to deliver an additional $20 million-plus of revenue to hit any version of the originally-implied $50 million 2026 target. The math is not impossible. It requires the second customer to convert at least $15 million of LOI to firm POs in the next six months, which the company has not publicly confirmed.
The ATM facility remains outstanding. If the second-half ramp slips, the company will need to raise additional capital before mid-2027. The clean balance sheet today does not mean the company is free of dilution risk through the back of 2026 and into 2027. The dilution risk is not the $250 million catastrophe scenario the 2024 bears warned about, because the post-rights-offering share count is already at 25 million and the equity base is $48 million. The dilution risk is the $5 million to $10 million capital raise at a depressed price if the ramp slips. That is the bear case range, not the apocalypse.
Customer concentration is the third risk. Two named LOI customers account for the bulk of the 2025 and 2026 revenue ramp. If either customer pauses the rollout, the revenue line slips by the same magnitude. Diversification is occurring slowly through the BEAD-funded regional carriers, but the named Tier 1 pair carries the revenue scenario in the base case.
Competitive risk is the fourth. The Asian ORU vendors will continue to bid on US deployments where the Buy America rules do not strictly apply, and the incumbent Ericsson and Nokia stacks could decide to ramp dedicated small-cell ORU products if the BEAD funding allocation forces them to. AmpliTech's small-cap advantage on speed and customization could erode at the volume tiers.
XI. Scenario Math
I do not run probability-weighted bear, base, bull scenarios with hand-tuned weights for a Part 1. I run scenario math that lets the reader weight it themselves.
Three revenue paths from a 2026 starting point:
Three sensitivity prices using a 4x EV-to-forward-sales multiple, which is below the historical RF small-cap median and roughly half the multiple Anokiwave commanded when Qorvo acquired it:
The bear-case multiple-and-revenue combination produces a stock 22 percent below today's $5.04 print. The base case is 57 percent above. The bull case is 182 percent above. The expected-value math, even on a flat 33-33-33 weighting across the three scenarios with no quantum optionality included, lands at roughly $8.70 against $5.04 today. The probability-weighted upside is 73 percent on a one-to-two-year horizon, before any contribution from the cryo-LNA line.
I am not pricing the quantum line. The quantum line is a tail. If the IBM 100,000-qubit timeline pulls into 2030 from 2033, the cryo-LNA revenue line alone exceeds today's total company revenue. That is not in the scenario math because I cannot probability-weight an IBM roadmap. It is in the optionality.
XII. The Catalyst Calendar
Eight named catalysts on the calendar through Q1 2027:
The first read is in mid-August. Q2 2026 results will show whether the second-half ramp is starting, whether gross margin is holding above 45 percent, and whether the LOI conversion has accelerated. That single print resets the multiple either direction.
The second read, harder to date, is the named-customer disclosure. The Tier 1 carrier the unnamed $78 million LOI sits with will eventually publish a BEAD or commercial deployment announcement that names AmpliTech as the radio vendor. That disclosure is what pulls a generalist tech analyst into the name and tightens the small-cap discount.
XIII. Kill Criteria
I run every position against an explicit kill-criteria list. The position exits if any of these trips:
One. The Q2 2026 print delivers revenue below $4 million. That would indicate the back-half loading is a stall, not a ramp.
Two. Gross margin reverts below 35 percent on a sequential basis. That would indicate the integrated-systems mix shift is reversing.
Three. The company files an 8-K disclosing a second material capital raise above $10 million at a price below $4.00. That would signal the ATM facility is being tapped under duress.
Four. Either named Tier 1 LOI customer issues a public statement pausing, descoping, or withdrawing from the AmpliTech procurement relationship.
Five. Maqbool sells more than 100,000 shares on the open market in any 90-day window. The historical pattern is zero open-market sales since IPO; a break in that pattern is a signal in itself.
Six. The IBM Quantum 100,000-qubit roadmap is publicly walked back, replaced with a non-superconducting architecture, or replaced with a different LNA vendor at the readout layer.
Seven. A new short writeup from a credible source publishes a thesis that documents specific financial misstatements, customer-relationship misrepresentations, or undisclosed contract cancellations. Existing bear writeups based on the 2024 guide miss and the BTC custody loss are already priced in.
None of those seven criteria are met today. None of them are imminent. The first testable one is Q2 2026 results in mid-August. Three months from today.
XIV. Position Plan
I have no AMPG position as of today, May 25, 2026.
I will open a starter position via daily DCA buys on the following schedule:
The four-tranche structure averages me into a name with 16 percent daily float turnover. Any single-session shock prints fast in this float, in either direction. DCA into the close on each day reduces single-print risk to one-quarter of the position size.
Target starter weight is small relative to the core book. I am not framing AMPG as a top-five position. It sits behind MRLN, IQE, BMM, LPKF, and the rest of the defense and chokepoint book in conviction order. AMPG is a tradable thesis, not a multi-year compounder anchor, until the Q2 print confirms or breaks the gross-margin step-up.
Conditional add line: a print below $4.25 on a non-news session adds one additional tranche from stable holdings. A print above $7.50 before Q2 earnings holds the position size flat. I am not adding above $7.50 pre-print.
XV. Bottom Line
AmpliTech Group printed a quarter on May 13, 2026 that the tape did not parse. Revenue up 48.6 percent year over year. Gross margin up 1,500 basis points. Cash $18.4 million. Debt zero. Stockholders' equity $48.4 million. Working capital ratio 4.25 against 1.68 three months earlier. The mix shift from parts-house to integrated 5G radio OEM showed up in the segment line and the margin line for the first time.
The pivot is real. The customer bench is real. The quantum optionality is real. The CEO is founder-aligned at more than 10 percent of the company, has never sold since IPO, and bought into the rights offering at the pro-rata. The bear case from the 2024 cycle is real and recent. The $2.75 print of late 2024 happened. The credibility problem is documented.
What the tape has not yet absorbed is that the unit economics of an integrated ORAN ORU at 48 percent gross margins is a fundamentally different business than the unit economics of a custom microwave parts house at 33 percent gross margins. The Q1 print is the first quarter where the math actually shows it.
The expected-value math on a one-to-two-year horizon, weighted flat across bear, base, bull at 4x EV-to-forward-sales, lands at $8.70 against $5.04 today. The quantum tail is not in that number.
The first testable catalyst is the Q2 2026 earnings print in mid-August. The DCA accumulation window for my starter position runs Tuesday May 26 through Friday May 29, 2026.
If the Q2 print delivers revenue at $7 million or above with gross margin at or above 45 percent, this becomes a sized-up position. If it delivers revenue below $4 million or gross margin below 35 percent, this becomes a closed position and the kill-criteria framework stops the thesis where it should.
I will report the average basis after the DCA window closes Friday.
Shawarma Capital
More in this series
$AMPG: The First Conversion. FCC And ISED Cleared The Radio, And The BEAD M
AMPG Stock: AmpliTech’s 48% Gross Margins, Open-RAN Radios, and a $7 Target
Great write up, we are also owning this for over a year now.