Are Marvell Lumentum acquisition jobs at risk?

What did Marvell announce, and does it affect jobs right now?

On November 2, 2026, Marvell agreed to buy Lumentum in an all-stock deal worth about $115 billion including debt, one of the largest acquisitions the tech industry has seen. Your job won't change this week. The deal still needs shareholder and regulatory approval, and it isn't expected to close until the first half of 2028. Lumentum shareholders will get 4.5 Marvell shares for every share they own, which leaves them with about a third of the combined company. Lumentum's CEO, Michael Hurlston, and two other Lumentum directors will join Marvell's board. Until the deal closes, the two companies have to run as separate businesses. That means no merged org charts, no shared roadmaps and no deal-driven layoffs yet.

The strategy makes sense if you work in optics. Marvell already sells the electrical side of data-center interconnect: the DSPs, drivers and transimpedance amplifiers inside pluggable transceivers. Much of that came from its 2021 purchase of Inphi. Marvell was also already buying lasers from Lumentum. Lumentum adds the photonic side: indium phosphide lasers, optical transceivers and optical circuit switching. Nvidia is a strategic investor in both companies after putting $2 billion into each earlier this year. AI clusters now need more bandwidth than copper can carry, so owning the laser, the DSP and the module together is a strong position against Broadcom. Lumentum's capacity has reportedly been sold out into 2029, so demand isn't the issue.

For employees, a different number matters most. The companies expect more than $1 billion in annual run-rate cost synergies. Some of that will come from purchasing, manufacturing sites and shared tooling. But in almost every large semiconductor merger, a real share of the savings comes from headcount in functions that now exist twice. That doesn't mean your role is on a list. Nobody has published one, and anyone who claims to know the cuts today is guessing. It does mean the next eighteen months are a window to plan. You can let that time pass, or you can use it to build a CV that would get you interviews elsewhere if the integration plan leaves your team out.

Which teams usually get cut after a chip merger like this?

Support teams that exist at both companies and overlapping product teams usually get consolidated first. Rare technical specialists are usually kept. In deals like this one, the first round of Marvell acquisition layoffs typically hits corporate functions: finance, HR, IT, legal, investor relations, and the sales and marketing teams that cover the same hyperscaler accounts. Those cuts often come soon after closing, because the overlap is obvious and the savings are easy to count. Engineering cuts are slower and more selective. Closing the wrong lab can stall a product line customers have already committed to buying, so integration teams usually spend months comparing roadmaps before they touch R&D headcount.

On the technical side, look for overlapping work rather than matching job titles. Both companies have worked on optical modules, silicon photonics and the test setups that qualify them. Where two groups are building competing 800G or 1.6T transceiver designs, chasing the same co-packaged optics roadmap, or running parallel test and characterization labs, the combined company will probably pick one. The program managers, product marketers and applications engineers tied to the losing roadmap are exposed, and so are the validation engineers who support it. Lumentum's laser chip design and indium phosphide fab work, on the other hand, is what Marvell is paying for. The people who can make high-speed EMLs and CW lasers at good yield are the reason for the premium.

Here's the part many people won't like: being 'core' to the deal protects you less than it sounds. Even teams that survive integration often get new managers, moved sites, rewritten roadmaps and a retention package. The vesting cliff on that package often ends up being the date people leave. Lumentum layoffs, if they happen, could also come through attrition and reorganizations rather than one big announcement. Your risk depends less on the logo on your badge today and more on whether your specific work is duplicated at the other company. If you can name a team there doing roughly what you do, assume your role will be reviewed and prepare for it.

What signals should engineers watch before the deal closes?

Watch four things: regulatory milestones, retention offers, hiring freezes and the integration org chart. Each one tells you how close consolidation is. Regulatory review comes first. A deal this big will be examined in several countries, and the closing date can move. If regulators require remedies, such as selling off a product line, that tells you which teams may end up with a different owner entirely. Next, watch hiring in your own company. If open roles in your group are frozen or quietly pulled while the deal is pending, leadership may be avoiding new hires it expects to cut later. That doesn't prove cuts are coming, but it's the earliest sign most employees ever get.

Retention bonuses are the most misread signal. A targeted retention offer means the company wants you to stay through closing and integration. It doesn't promise you a job after the payout date. If people on a matching team at the other company got one and you didn't, that tells you something. Once an integration office is set up, notice who leads the workstream for your product area and which site they work at. Leaders tend to consolidate around their own sites and tools. Other things to track: your project's priority in quarterly planning, customers asking which roadmap will survive, and tool or PLM migrations that show whose systems are winning.

None of these signals is a reason to panic, but each one should move your CV update from 'someday' to 'this month'. Job security during an acquisition mostly comes down to having options. Engineers with a current, targeted CV can make calm decisions when a retention offer or a reorg arrives, instead of scrambling. Updating your CV is also private. You don't have to tell anyone, apply anywhere or take a recruiter's call. It just means that when a contact at Coherent, Ciena, a hyperscaler's hardware group or an AI accelerator startup asks if you'd consider a move, you can say yes and send something that evening instead of three weeks later.

What should an optical engineer resume show in 2026?

In 2026, an optical engineer resume should lead with data rates, device types and measurable results: what you designed, characterized or shipped, at what speed, and what it improved. Hiring managers for AI-infrastructure optics look for terms like 800G and 1.6T, PAM4, EML, CW laser, silicon photonics, co-packaged optics, linear pluggable optics and coherent modules such as 400ZR. They also look for test terms: BER, TDECQ, eye diagrams, sensitivity, reliability qualification and yield. Applicant tracking systems do the same thing, only more literally. If your CV says 'worked on next-gen transceivers' instead of '1.6T OSFP transceiver bring-up, 8×200G PAM4', both the recruiter and the software miss the match. Internal code names mean nothing outside your company, so translate every one.

A strong photonics engineer resume for 2026 also makes your technical skills easy to find. Give them a plain section called Technical Skills, split into lines: devices (EML, DFB, VCSEL, CW lasers, modulators), platforms (InP, silicon photonics, SiN), design and simulation tools (Lumerical, RSoft, Zemax, MATLAB, Python), and test equipment (sampling oscilloscopes, BERTs, optical spectrum analyzers). Keep the layout to one column, avoid tables and text boxes, and save it as a text-based PDF or Word file. Then put a number in every bullet: how much wafer yield rose, how much test time per module fell, how much TDECQ margin you gained, whether qualification finished on schedule. Numbers are what let a hiring manager outside your company understand what an internal win was worth.

Test, validation and product engineers need to be especially careful here, because their work is the easiest to undersell. 'Performed testing on optical modules' tells a recruiter nothing. 'Automated a TDECQ and sensitivity test flow in Python across 12 stations, cutting per-unit test time by 40%' shows you know the measurements, the scale and the result. Just use your real figures. The same goes for manufacturing and process engineers: name the process, the yield metric and the volume. If you've worked with hyperscaler customers, say so plainly without breaking confidentiality. 'Supported qualification with three tier-one cloud customers' is safe, specific and exactly what other AI-infrastructure employers want to see.

How do you retarget your CV at other AI-infrastructure employers?

Build one master CV, then tailor a version for each job ad using that employer's own wording. Optics hiring outside this deal is broad. It includes transceiver and laser makers, networking chip companies, GPU and accelerator vendors adding optical links to their systems, hyperscalers' in-house hardware teams, and startups working on co-packaged optics and optical I/O. Each one describes the same skills differently. A hyperscaler ad may ask for 'link qualification' and 'fleet reliability', while a component maker asks for 'device characterization' and 'yield'. Your experience fits both, but the words in your CV need to match the ad you're answering, or the first screen won't make the connection. Start with a master list of everything you've done, then cut it down for each role.

The process itself is simple. Put the job ad next to your CV and mark every tool, device type, data rate and test method it names. Then check that each one appears in your CV in the same form. If the ad says 'silicon photonics' and you wrote 'SiPh', spell it out. If it asks for Python test automation and you've done it but never listed it, add it. To speed this up, you can check whether your CV matches a specific job description with HRLens. Paste the text of the ad (not a LinkedIn or Indeed link, since those sites block fetching) and upload your CV. The analysis lists the skills the ad asks for that your CV doesn't show, along with a score out of 100 and an ATS compatibility rating.

If you haven't touched your CV since you joined Lumentum or Marvell, rebuilding it is usually faster than patching it. Start in the HRLens CV builder: paste your old CV or describe your work in the chat, then refine each bullet until it names a device, a data rate and a result. Pick one of six ATS-friendly templates. When it's done, run it through the free CV analysis to see your overall score and five category scores, including impact and ownership and ATS compatibility. The paid Full Analysis adds a fix list and the likely reasons you'd be rejected. Do this now, while you're employed and the deal is still about eighteen months from closing. That's the strongest position you'll be in.

Frequently asked questions