Software Engineer | Together AI Global Negotiation Guide

Negotiation DNA: Pre-IPO Equity + Base + Bonus | AI Inference & Optimization | 400% YoY Growth | $3.3B Valuation | +20% Kernel Innovation Premium

Region	Base Salary	Equity (Options/RSU est.)	Bonus	Total Comp
San Francisco	$145K–$185K	$30K–$55K	10–15%	$165K–$228K
Seattle	$138K–$176K	$28K–$52K	10–15%	$157K–$217K
Remote US	$130K–$166K	$25K–$48K	10–15%	$148K–$205K

Negotiation DNA

Together AI is the company behind FlashAttention — the single most impactful algorithmic optimization in modern AI inference, now used by every major AI lab from OpenAI to Google DeepMind. As a Software Engineer at Together AI, you contribute directly to the Together Kernel inference optimization framework that powers this 400% YoY revenue growth engine. Your code ships into the foundational infrastructure layer that makes AI deployment fast and affordable across the entire industry, giving you leverage that few early-career engineers at any company can claim. (Source: Together AI Series A announcement, Forbes 2025; Crunchbase funding data.)

Level Mapping: Together AI SWE ≈ Google L3–L4 · Meta E3–E4 · OpenAI IC1–IC2 · Anthropic SWE I–II · Amazon SDE I–II

Together Kernel — The FlashAttention Premium

Together AI created FlashAttention — the algorithm that redefined how every transformer model handles attention computation, cutting memory usage and latency by orders of magnitude. Together Kernel is the proprietary inference optimization framework built on top of this breakthrough, delivering state-of-the-art speed and efficiency that no competitor has matched. With 400% YoY revenue growth and a $3.3B valuation, Together AI is on a rocketship trajectory fueled by kernel-level innovation. As a Software Engineer, you are building the optimization layer that makes AI inference fast and cheap everywhere — the foundational infrastructure of AI deployment. This justifies a 20–30% Kernel Innovation premium over comparable roles at other startups. Frame your negotiation around the fact that your work doesn't just serve Together AI — it powers inference pipelines industry-wide, and that kind of impact commands premium compensation.

Global Levers

1. FlashAttention Ubiquity: "My work contributes to the inference optimization stack that every major AI lab depends on — FlashAttention is the industry standard, and I'm building on it at the source."
2. Pre-IPO Equity Upside: "At a $3.3B valuation with 400% YoY growth, my equity has asymmetric upside — I need the strike price and vesting schedule to reflect that I'm taking real pre-IPO risk."
3. Kernel Scarcity: "Engineers who can work at the kernel and systems optimization layer are extremely scarce — the intersection of CUDA, memory optimization, and ML inference is a talent market of hundreds, not thousands."
4. Competing Offer Leverage: "I have competing offers from [FAANG/top AI lab] at $X total comp — Together AI's offer needs to close the gap with equity that reflects the growth trajectory."

Negotiate Up Strategy: "I'm excited about Together AI's mission and the chance to work on the inference stack behind FlashAttention. Based on my competing offer of $210K TC from [Company], I'm looking for a base of $180K with an equity grant valued at $50K annually at current 409A, plus a 15% target bonus. My accept-at floor is $195K total comp — below that, the risk-adjusted math doesn't work against my FAANG alternative. I'd also like to discuss a sign-on bonus of $15K–$25K to offset my unvested equity elsewhere."

Evidence & Sources

• Together AI $3.3B valuation and funding history (Crunchbase, 2025)
• FlashAttention paper citations and industry adoption (Dao et al., 2022; arXiv)
• Levels.fyi AI startup compensation benchmarks (2025–2026)
• Glassdoor and Blind salary reports for AI infrastructure startups (2025–2026)