Company wants me to pay back "signing bonus" that was actually just regular pay?
I started at this marketing agency back in August and during negotiations they mentioned a $3000 signing bonus.
Thing is, they never paid it as a lump sum. Instead my first three paychecks were about $1000 higher than my actual salary should've been. I thought it was just them spreading it out or whatever and didn't think much of it since I was busy with the new job, plus I already had some money saved aside so I wasnt really tracking every dollar.
Fast forward to now and my manager pulls me into a meeting saying I need to pay back the "signing bonus" because I didn't stay a full year (im at 5 months). But here's the thing, there was never any paperwork about this bonus or repayment terms. Nothing in my offer letter mentions it either, it just lists my base salary. When I brought this up they said it was "verbally agreed upon" during the interview but I honestly dont remember that conversation happening at all.
They're saying they're gonna deduct it from my final paycheck when I leave next month. Can they even do this without any written agreement? The whole thing feels off because those paychecks just looked like regular pay, not bonus pay.
Jobadvisor
That sounds incredibly frustrating and, frankly, highly irregular. It feels "off" because most companies have very strict protocols for bonuses to protect themselves—and you—from exactly this kind of confusion.
In many jurisdictions, what they are attempting to do is legally shaky at best. Here is a breakdown of why this situation is problematic and how you might want to handle it.
1. The "Paper Trail" Problem
A signing bonus is a contract. For a company to legally enforce a "clawback" (making you pay it back), there almost always has to be a written agreement signed by you that outlines:
The exact amount of the bonus.
The required length of service (the "retention period").
The specific terms for repayment if you leave early.
Without a signed document or a clause in your offer letter, a "verbal agreement" is very difficult for an employer to enforce. In many states and countries, labor laws strictly prohibit employers from deducting money from a final paycheck unless it is for taxes, court-ordered garnishments, or something the employee specifically authorized in writing.
2. Bonus vs. Wages
The way they paid you is a major red flag.
Bonuses are usually coded differently in payroll systems for tax purposes (often subject to a flat supplemental withholding rate).
Regular Wages are just that—pay for hours worked.
If they simply bumped up your first three paychecks and those stubs list the money as "Regular Earnings" or "Salary" rather than "Bonus," they may have a very hard time proving in court (or to a Department of Labor) that this wasn't just your agreed-upon salary for those months.
3. Potential Next Steps
Gather Your Documents: Download every single pay stub you have from August to now. Look closely at the "Earnings" descriptions. Also, save a copy of your signed offer letter and any employee handbook you were given.
Check Your Offer Letter (Again): Look for phrases like "other compensation" or "discretionary payments." If it truly only lists your base salary and nothing else, you have a strong hand.
Request Proof: Ask your manager or HR for the signed document where you agreed to the repayment terms. If they say "it was verbal," you can calmly respond: "I have no record or recollection of a repayment agreement, and my offer letter does not outline these terms. Without a signed authorization, I do not consent to a deduction from my final paycheck."
Contact Your Local Department of Labor: Most locations have a labor board that handles "wage theft" or "unauthorized deductions." They are usually very helpful and can tell you exactly what the laws are in your specific area.
A Note on "Final Paycheck" Laws
In many places, the law requires the final paycheck to be paid in full on your last day (or shortly after). If they withhold $3,000 without your written consent, they could be liable for "waiting time penalties," which might end up costing them much more than the $3,000 they're trying to get back.
Should I just accept the voluntary termination?
Hello HR. I work retail as an associate. I posted a few days ago about being on personal leave for surgery recovery. Not eligible for FMLA because I don't meet the hours requirement so I am on personal leave for 4 weeks till this Saturday. They said this last Monday:
Since you are unable to return without clearance, we will be recommending termination effective 1/17/26 as a voluntary termination due to not being able to return from leave. This termination type is eligible for rehire, as positions are available, and once you are able to provide clearance from your doctor's to (HR email here).
I was thinking about getting a note from my doc to avoid termination but I don't think he will agree to oblige until 8 weeks are up. I am approaching the 4 week recovery mark.
The "as positions are available" part, does it imply that positions are available and I can simply send a note when I am able to work to get reinstated? Or is it a sort of "when positions are available" sort of circumstance?
My store hardly hires new people and I do plan to ask about the rehiring process tomorrow. I've been here 10 years but I have never seen anyone in my situation. I did apply for EDD and still waiting. What is your interpretation? Thank you
Jobadvisor
This is a stressful situation, especially after 10 years with a company. In California, the laws are generally more protective of employees, but the "voluntary termination" they are proposing is a specific administrative tactic.
Here is a breakdown of what is likely happening and why you should be cautious.
1. The "Voluntary" Label vs. Reality
In California, calling a termination "voluntary" when you actually want to keep your job (but are physically unable to work) is a bit of a legal gray area.
The Good: By labeling it "voluntary" and "eligible for rehire," they are signaling they don't have a "grudge" against you. It preserves your reputation for future background checks.
The Bad: The phrase "as positions are available" is a huge caveat. It means they are not holding your specific spot. If you are cleared to work in four weeks and they’ve already hired someone else to cover your shifts, they have no legal obligation to bring you back immediately. You would essentially have to re-apply and wait for an opening.
2. Unemployment Benefits (EDD)
This is where the "voluntary" label gets tricky. To collect unemployment in California, you generally must be unemployed through no fault of your own.
If the paperwork says you "voluntarily quit," the EDD may initially deny your claim.
However, California is usually sympathetic if you can prove you left because of a medical necessity and intended to return.
Important: You cannot collect Unemployment Insurance (UI) while you are still recovering, because you must be "able and available" to work. You should be on State Disability Insurance (SDI) while you are recovering, and then switch to UI only once your doctor clears you but your job is no longer there.
3. California-Specific Protections (CFRA)
You mentioned you aren't eligible for FMLA, but have you checked the California Family Rights Act (CFRA)?
In California, if your employer has 5 or more employees, you are protected under CFRA if you have worked 1,250 hours in the last year.
If you don't meet the hours, you may still be protected under California’s Pregnancy Disability Leave (PDL) (if applicable) or as a Reasonable Accommodation under the Fair Employment and Housing Act (FEHA).
Since you have 10 years of seniority, a request for an extra 4 weeks of leave is often considered a "reasonable accommodation" in California, rather than firing you.
4. My Interpretation of Their Offer
They are essentially saying: "We aren't going to fire you for misconduct, but we aren't going to wait for you either. We are taking you off the books now to stop your benefits/seniority accrual, and if we have a hole to fill when you're healthy, we might put you in it." It is a "soft" termination that benefits the company more than you because it avoids a messy "firing" on their record.
Suggested Next Steps
Don't Sign Anything Yet: If they ask you to sign a "Resignation Letter," be very careful. That could hurt your ability to get disability or unemployment.
Request an "Extension of Leave as an Accommodation": Before accepting termination, send an email to HR:
"I have been with the company for 10 years and value my role here. While I don't qualify for FMLA, I would like to request an additional 4 weeks of leave as a medical accommodation under FEHA. I intend to return as soon as I am cleared."
Ask for Reinstatement Specifics: When you talk to them tomorrow, ask: "If I accept this, will I retain my 10 years of seniority for pay and vacation purposes if I am rehired?" (Usually, the answer is no, which is a big loss for you).
Internships shouldnt be restricted to just students. Cant find new grad jobs
I have been software engineer for ~5 years, in IT for 8. Just graduated with masters in CS, but I screwed myself by not being able to get an internship (I did try, nobody got back to me).
Now I have no experience outside of student projects. While some are very creative and outside the box, and not your typical project that screams "Proof of concept", I have no actual experience with machine learning or computer vision past school. Every entry level job seems like they want you to have "strong foundational skills", "works with tensor flow, pytorch, published working applications... some have the classic "1-2 years experience"... like I would gladly accept crappy wage to have the company willing to teach me the ropes so to speak. Why cant people get internships just to learn it makes no sense, im working for free or for extremely low wages.
I hate that universities dont teach these practical skills, I'd gladly do a mock work environment where you just make a bunch of projects using common skills asked.With Software engineering I got really lucky and applied internally. I was such a beginner that looking back I had no idea what an actual production environment was like, my projects were so simple
Jobadvisor
It is incredibly frustrating to feel like you’ve done everything "right"—getting a Master’s degree and putting in 8 years of IT work—only to hit a wall because of a lack of a specific internship title.
The reason you’re seeing these restrictions isn’t just "company policy"; it’s largely due to labor laws. In many places (especially California), the legal test for an unpaid internship is that it must be for the primary benefit of the student and often must be tied to academic credit. If a company hires a non-student and doesn't pay them, they risk being sued for wage theft because that person is legally an "employee," not a "trainee."
However, with 5 years of Software Engineering (SWE) experience and a Master's, you are actually in a much stronger position than a "new grad" with no work history. You just need to pivot your strategy from seeking internships to seeking bridge roles.
1. The "Experience" Paradox
Companies asking for 1–2 years in Machine Learning (ML) or Computer Vision (CV) for "entry-level" roles are often looking for demonstrated proficiency, not necessarily a specific job title.
Your Advantage: You already know how production environments work. You understand CI/CD, version control, and technical debt—things 22-year-old interns don't know.
The Gap: You need to show that you can apply your SWE rigor to ML/CV frameworks ($TensorFlow$, $PyTorch$, $OpenCV$).
2. Strategy: The "Internal Transfer" or "Adjacent Role"
Since you’ve already successfully navigated an internal move once, consider these paths:
MLOps Engineer: This is the perfect middle ground for you. It requires 80% SWE skills (infrastructure, deployment, scaling) and 20% ML knowledge. Companies are desperate for people who can actually deploy models, not just build them.
Computer Vision (CV) Integration: Look for roles in companies that do "Applied AI." They often need a solid Software Engineer to help their researchers integrate CV models into the actual product.
3. Creating Your Own "Mock Environment"
If the university didn't teach the practical side, you can build a "Production-Ready" portfolio that looks like real work. Instead of a school project, build a Full-Stack ML App:
Data: Scrape or use a public API.
Model: Train a model using $PyTorch$ or $HuggingFace$.
Deployment: Containerize it with Docker, deploy it to AWS/GCP, and set up an API endpoint (FastAPI).
Monitoring: Add a simple dashboard showing model drift or latency.
Note: A recruiter seeing a deployed, monitored ML system on GitHub will value that far more than an internship where a student just cleaned data for 3 months.
4. Legal Loophole: "Fellowships" and "Returnships"
Since you find the internship door closed, look for these specific terms:
Fellowships: Programs like Insight Data Science or The Data Incubator are designed specifically for people with advanced degrees (like your Master's) to transition into ML.
Apprenticeships: Companies like Google, LinkedIn, and Microsoft have apprenticeship programs that are often open to non-students or career changers.
Comparison of Roles for Your Background
| Role | Why you fit | The "Learning" Gap |
| MLOps Engineer | 8 years IT/SWE experience is perfect for infrastructure. | Learning $Kubeflow$, $MLflow$, and model monitoring. |
| Computer Vision SWE | You can write the "wrapper" code for CV algorithms. | Basic $C++$ or $Python$ CV libraries ($OpenCV$). |
| Data Engineer | You likely already know SQL and Backend systems. | Learning Spark, Airflow, and Big Data pipelines. |
Next Step
Since you are in California and have a 10-year history with your current employer, have you checked if your current company has an ML or Data Science team you could "shadow" or transition into? Given your seniority, they might be more willing to let you "learn the ropes" than a stranger would.
Has personalized job matching actually helped anyone, or is it just a fancy way to waste time
I keep seeing job boards brag about personalized job matching like it’s going to save my life. But in reality it feels like my feed is just the same 10 jobs wearing different hats.
It keeps recommending roles that either don’t match my experience at all, are wildly underpaid, are in the wrong industry or are things I clicked once out of curiosity and now I’m trapped forever
The funniest part is it’s supposed to reduce doom scrolling, but I feel like I’m doom scrolling more just guided by an algorithm with confidence.
I’m genuinely curious if anyone has actually gotten interviews or hired through personalized job matching, or if this is just a fancy way to keep people on the platform longer.
Jobadvisor
It is definitely not just you. What you’re describing is a well-documented phenomenon in the industry: the "echo chamber" of algorithmic job matching.
While these systems are designed to help, they often fail because they are built on "Keyword & Behavior" logic rather than "Career Intent" logic. Here is the reality of why it feels like a waste of time and whether anyone actually wins at this game.
Why it feels like "Doom Scrolling with Confidence"
The algorithms on sites like LinkedIn and Indeed primarily use three data points, all of which have massive flaws:
The "Clicked Curiosity" Trap: If you clicked a "Flight Attendant" job once because the photo looked cool, the algorithm marks you as "Interested in Aviation." Because it wants to keep you on the platform, it will flood you with similar roles to get another click—even if you're a Software Engineer.
The Keyword Loop: Most matchers are just advanced Ctrl+F. If your resume says "Project Management" and a job for "Construction Site Manager" also says "Project Management," the AI thinks it's a 100% match, ignoring the fact that you work in Tech and have never worn a hard hat.
The "Freshness" Bias: Platforms prioritize showing you new jobs to make the site feel "active." This is why you see the same 10 roles "wearing different hats"—they are the ones being aggressively promoted or reposted to stay at the top of the "Personalized" feed.
Does it actually work for anyone?
The short answer: Yes, but mostly for "Standard" roles.
Personalized matching works best for people with very linear, common career paths. If you are a "Retail Store Manager" looking for another "Retail Store Manager" role, the algorithm is actually quite good at finding those.
Success Stories: People who get hired through these features usually treat them as a notification service rather than a search tool. They set very specific filters and only engage with the "Top Matches" that actually align.
The "Easy Apply" Paradox: Roles that show up in your "Personalized" feed often have 500+ applicants because everyone saw them in their feed. Your "match" might be high, but your "competition" is also at an all-time high.
How to break the loop (and actually get an interview)
If you want to stop the "guided doom scrolling," you have to "retrain" the AI or bypass it:
| Tactic | Why it works |
| Clear your "Interests" | Go into your settings (especially on LinkedIn) and delete the job titles/locations you once clicked on out of curiosity. |
| The 24-Hour Rule | Only look at jobs posted in the last 24 hours. The "Personalized" feed often pushes older, "stale" jobs that the platform is struggling to fill. |
| Boolean Search | Stop using the "For You" tab. Use the search bar with operators like “Product Manager” NOT “Senior” to force the algorithm to be precise. |
| Go to the Source | Use the job board to find the company, then go directly to the company’s "Careers" page to apply. This bypasses the board's internal "ranking" system. |
The "Fancy Waste of Time" Verdict
Personalized matching is 50% utility and 50% engagement hacking. It is designed to make you feel like the "perfect job" is just one more swipe away.