When Elon Musk walked into Twitter HQ in October 2022, one of the first names on the chopping block was Parag Agrawal — the CEO who had been on the job for less than a year. Three and a half years later, that same Parag Agrawal is back in the spotlight, but on his own terms. His new AI startup, Parallel Web Systems, has just hit a $2 billion valuation, and almost no one saw him coming.
From Twitter's corner office to the unemployment line
Agrawal's Twitter story was supposed to be a fairy tale. He joined the company as a software engineer in 2011, rose through the ranks to become CTO, and was personally tapped by Jack Dorsey to take over as CEO on November 29, 2021. Less than a year into the job, his world flipped. Musk closed his $44 billion takeover, and Agrawal was shown the door within hours — reportedly walking away with a $42 million severance package.
While most fired CEOs run the speaking-circuit-and-board-seats playbook, Agrawal disappeared. He spent 2023 quietly assembling a small team of engineers and researchers and started building something most people hadn't even realized was a problem yet: infrastructure for AI agents to actually use the web.
Enter Parallel Web Systems
Parallel, founded in early 2024 and headquartered in Palo Alto, builds the plumbing autonomous AI agents need to search, read, and act on the web. Agrawal's bet is simple but bold: as AI agents start handling research, legal discovery, claims processing, and other heavy workflows, they'll consume the web at a scale humans never could — and they'll need tools built for machines, not browsers built for people.
The market clearly agrees. Here's what the funding trajectory looks like:
November 2025 — Series A of $100 million, valuing the company around $740 million
April 2026 — Series B of $100 million led by Sequoia Capital, valuation jumps to $2 billion
Total raised so far — roughly $230 million
Other backers — Kleiner Perkins, Index Ventures, Khosla Ventures, First Round Capital
Going from a Series A to a $2 billion valuation in about five months is one of the fastest valuation jumps in this AI cycle. And it happened while the company was barely talking publicly — Parallel only emerged from stealth in late 2025.
Quick Facts: The Twitter–Musk–Agrawal Timeline
For readers who want the receipts, here's the timeline at a glance:
November 29, 2021 — Parag Agrawal becomes CEO of Twitter, replacing Jack Dorsey.
April 14, 2022 — Elon Musk makes an unsolicited $44 billion offer to buy Twitter.
April 25, 2022 — Twitter's board accepts the offer.
July 2022 — Musk tries to back out of the deal, citing spam bot concerns. Twitter sues.
October 27, 2022 — Musk officially closes the $44 billion acquisition. Agrawal is fired the same day, along with CFO Ned Segal and legal head Vijaya Gadde.
November 4, 2022 — Musk lays off roughly 3,700 employees in a single day — about half of Twitter's ~7,500-person workforce.
Mid–late November 2022 — Hundreds more resign after Musk's "extremely hardcore" ultimatum.
By April 2023 — Headcount drops from around 8,000 to under 1,500, a roughly 80% reduction.
July 23, 2023 — Twitter is rebranded as X.
April 2026 — Parallel Web Systems hits a $2 billion valuation under Agrawal's leadership.
Why this story matters
There's something poetic about the arc. The CEO who was publicly humiliated by the world's richest man — and who became the punchline of countless memes about Twitter's chaotic transition — is now quietly building infrastructure for the next era of the internet. While Musk has been busy turning Twitter into X, fighting advertiser boycotts, and launching xAI, Agrawal has done the opposite: said almost nothing, shipped quietly, and let the funding rounds speak.
Asked recently about the pace at Parallel, Agrawal kept it characteristically understated: every few weeks the team solves one bottleneck and runs straight into the next one.
That's the sound of a comeback that doesn't need a press tour.
The takeaway
Getting fired from a high-profile job isn't the end of a career — sometimes it's the start of a much more interesting one. Agrawal had every reason to take a quiet board seat or cash out into early retirement. Instead, he picked one of the hardest problems in AI, built it from scratch, and turned a brutal exit into a $2 billion second act.
If AI agents really do become the way most of the web gets used, Parallel Web Systems may end up being far more consequential than the platform Agrawal was famously fired from.