The big-company offer: levels and equity

The loop produces a level recommendation, and it comes from scope evidence, not from years. Three things move it:

• The size of the thing you owned. "I built features on three applications" is one level. "I own architecture, release and reliability across three applications, and designed the service layer on a platform in four environments" is the next one. Both describe your job.
• Whether your decisions affected people outside your team. This is the SDE-2 to SDE-3 line almost everywhere. Setting a standard others followed counts; shipping a lot does not.
• How you answered system design and the behavioural rounds. A strong coding round rarely raises a level. A design round where you named trade-offs with numbers does.

So the level conversation happens in R8 and R10, hours before anyone says a number.

It happens for two reasons that look identical from outside: the loop did not produce the evidence, or the team has a headcount at that level and nothing above it. The first is about you and the second is not, and the recruiter will usually tell you which if you ask plainly.

• Worth accepting when the company is a level up from where you are, when the band at the lower level still beats your current number substantially, and when promotion cycles there are known and real. Getting into a strong company one level low and promoting in eighteen months is a common, good path.
• Not worth accepting when it comes with the same money as your current role, when the team cannot describe what promotion requires, or when "we will re-evaluate in six months" is offered instead of anything written.

A typical Indian big-company offer for six years of experience is three things:

• Base — paid monthly, taxed as salary. This is the only part you can plan around.
• RSUs — a grant of shares vesting over four years. Restricted stock, not options: you do not pay to get them, and they are worth the share price on the day they vest.
• Sign-on — cash, usually paid across the first one or two years, and usually there to fill the gap while the stock is still vesting slowly. Frequently clawed back if you leave inside a year.

The vesting schedule is the part that decides what a package is actually worth in year one:

Which means a "₹48 lakh" Amazon offer and a "₹48 lakh" Google offer can differ by several lakh in the first year, in Google\'s favour, and then reverse in year three. Compare year one and year four separately, and know which one you are optimising for.

• Listed company (Amazon, Google, Microsoft, Salesforce). The stock is liquid — it vests, you can sell it that day. Value it at today\'s price with a haircut for the fact that you are exposed to one company\'s share price for four years. Treating it as cash is roughly right; treating it as zero is throwing money away.
• Late-stage private (a unicorn, pre-IPO). Value it at the last round\'s price and then discount it hard. There is no buyer until there is a buyer, secondaries are rare, and the last round\'s valuation is a negotiated number, not a market price.
• Startup ESOPs. Assume zero for planning and negotiate on cash. If you are told otherwise, ask for the strike price, the total shares outstanding and the last preference stack — the answers are usually the end of the conversation.

The tax part matters and surprises people. In India, RSUs are taxed twice: as a perquisite at the moment they vest, at your slab, on the full market value; and then as capital gains on whatever the price does between vesting and selling. Foreign shares held over two years are long-term. Employers usually sell a slice at vest to cover the perquisite tax, which is why the number of shares that land in your account is smaller than the number that vested.

The initial grant runs out. What replaces it is the annual refresher: a new grant made each year at review time, also vesting over four years, which after two or three years stacks so that something vests every quarter. This is how big-company compensation actually compounds, and it is invisible in the offer letter.

It is also the honest answer to the Amazon back-loading question. Years three and four look enormous on paper because the original grant is 80% of it — but by then refreshers are also landing, which is why total compensation keeps climbing rather than falling off a cliff in year five.

Put both on the same footing before you feel anything about them. Four numbers, not one:

• Year-one cash — base plus sign-on plus whatever vests in year one, after the perquisite tax.
• Year-four run rate — base plus a full year of vesting, assuming the price does nothing.
• Guaranteed — the part that arrives even if the stock halves and you leave in eighteen months.
• Monthly in-hand — the number your life runs on, and the comparison R12 keeps making. It stays the right one at this level too, even when most of the package is paper.

A ₹32 lakh fully-fixed startup offer and a ₹45 lakh big-company package can be the same money in year one and very different money in year four. Neither is automatically better; they are different bets, and the deciding factor is usually not the money at all — it is whether you want the scope of a small company or the levelling ladder of a large one.

• Team matching — at Google and some others you can pass the loop and then wait for a team. It is normal, it takes weeks, and the offer is not real until it lands somewhere.
• Hiring committee — a group who never met you reviews the packet and decides. Your interviewers\' written notes are the entire input, which is R11·LP\'s whole point.
• Offer expiry — usually a week, occasionally "exploding" in 48 hours. Ask for time; a company that will not give you a few days to compare is telling you how it operates.
• Relocation and joining bonus are separate line items and are often the easiest thing to add when the base cannot move.
• Background verification at this size takes weeks and checks everything you claimed — dates, titles, and the CTC you stated in R1. This is why that number had to be true.