Cofounder, CSO @perplexity_ai. Former high frequency trader, competitive programmer. Think fast, build faster.
New York, NY
Joined December 2011
- Tweets175
- Following282
- Followers229K
- Likes2.4K
At Perplexity, agent prompts are no longer typed by humans: increasingly teammates are prompting each other, listening to automations, and ambiently listening to human feedback. Perplexity bridges the context between a large number of cloud tools and software services: searching and processing more than a human could possibly track manually. Hard decisions, restraint, and judgment before shipping are of course still enforced by humans.
A particularly important value at Perplexity is to always seek the truth. In today's stack, this means centralizing the ground truth in git repos and data warehouses when answering questions about the business, instead of looking at intermediate, often inaccurate or outdated artifacts. This is what enterprise cloud agents like Perplexity are especially good at: connecting the dots across your organization and monitoring the facts 24/7.
Even as an AI-native company it took us many months of development and cultural transformation to get to this point. It will be fascinating what the world looks like when all businesses reach this level of automation and beyond.
Today we are rolling out the latest form factor for work: Perplexity teammates. Agents have reached the point where they can start to be trusted with responsibilities at work. Teammates excel at automating tasks like deep research, data analysis, QA, triaging tickets, and reviewing code, which has saved each of us internally at Perplexity hundreds of hours of menial work. Leaving us the much more interesting work of setting vision, judging quality, and teaching restraint.
Agents are now rising up beyond tools and task-based sessions to take on supporting and coworking roles. Teammates proactively reach out to you, other coworkers, and other teammates in the workspace. Teammates close the loop on their own work, finishing the job instead of requiring repeated steering or manual management of sessions.
Available to Max users as a research preview today.
Automate ad agencies with Computer. Save time so you can focus on the hard problems that aren't yet solved by AI, like positioning and creativity.
Computer managed $30k in Meta ad spend for Get Maine Lobster, outperforming their agency.
Compared to their usual ads, Computer delivered 3.5x revenue with 63% higher ROAS. It automated creative analysis, ad production, and media buying.
Read more: pplx.ai/gml
This level of attention to detail is not yet solved by AI (for now).
Tell Claude to implement these 3 concepts to instantly make your platformer feel fantastic
Each one quietly forgives a mistake the player never notices making
1 Coyote time
Players often press jump a moment after running off a ledge. To them they were still on it. A basic controller ignores that press and they fall, and it feels like the game cheated. Let the jump still work for about 6 frames (0.1s) after leaving the ground
2 Jump buffering
When chaining jumps, players press jump just before they land. A basic controller throws that press away, so the character lands and stands still. Keep the press for about 6 frames and fire the jump on the frame they touch down
3 Corner correction
A jump that misses a ledge by a few pixels hits the wall and slides down, even though it looked like it made it. If the feet are 4 px or less under the ledge top, nudge the player up onto it. The same trick works for a head clipping a ceiling corner
Each one is under 10 lines of code but makes a huge difference
Props to @pika_labs for supporting the open weights ecosystem. Now the top open weight model for text to video.
Meet Pika H3 Flex, debuting today as the #1 open weight model on DesignArena’s t2v board.
It’s built for professional iteration, with exportable audio stems, resolutions up to 4k, and faster-than-realtime generation. It can produce a 15-second clip with soundtrack in as little as 7 seconds.
So, yeah. It’s kind of a flex 💅
The hard part of security work still requires human ingenuity.
Just dropped a new blog post! 🔥
blog.orange.tw/posts/2026-10…
Voice is the ultimate UI/UX for orchestration - spawn as many agents as you need all from one click.
An incredible result. Worth noting the great storytelling and presentation here to make the significance of the work understandable by everyone.
Impressive new results from Argon, now available in Perplexity. The frontier and cost/quality tradeoff is extremely close between the top models.
The sandboxes that power Perplexity are now open source.
It's been a lot of fun to be contributing a few spare commits in my spare time. Definitely would not have time to be actively learning a new language and rewriting our stack in Rust without today's coding agents and Perplexity's developer tooling.
We're open sourcing SPACE, the sandbox platform that runs every task in Perplexity Computer.
SPACE gives each agent task its own Firecracker microVM and controls credentials and outbound traffic from outside the VM.
Available now under Apache 2.0: github.com/perplexityai/SPAC…
Make hard decisions quickly, no matter how painful.
When I was a teenager, my friends and I came up with “the coin flip method” from first principles. If we were indecisive or were at a stand still on something, we’d flip a coin to make a hard decision knowing fully well if we felt bad about what the coin told us, we would go the other way. Like if we should rent movie A or movie B.
Later I learned this is called a “Freudian coin toss,” and knowing when you are displeased with a result is a strong signal for anything and a great way to keep motion.
Perplexity augments fast and cheap models like GPT 5.6 Luna to improve accuracy and overcome hallucinations. Now compatible with any open harness: via CLI, SDK, or MCP.
Perplexity Search debuts on the Artificial Analysis Search Index, with all three context size variants taking top positions on the leaderboard
The @perplexity_ai Search API comes with three context settings (low, medium, and high) that control how much extracted content each search result carries. We tested all three variants using our standardized methodology: the same model (GPT-5.6 Luna at medium reasoning), running inside Stirrup, our open-source agent harness, with tools for searching and fetching pages from the web. Only the provider behind the search tool changes.
Key results:
➤ Perplexity Search (medium) scores 80 on the Artificial Analysis Search Index, ahead of the previous leaders, Parallel (advanced) and Brave Search (LLM context), at 75. The high and low variants score 79 and 77 respectively. Its lead is concentrated in BrowseComp results, with AA-Omniscience and DeepSearchQA scoring comparably to other leading providers
➤ Efficient search payloads: smaller overall search results mean the model reads less per task, so Perplexity has the lowest model inference cost per task of providers we’ve tested so far, ranging from $0.028 to $0.034 across the three variants vs $0.036 for the next lowest provider
➤ Total cost per task is ~$0.091 for the medium and high context variants, at mid-pack latency. For comparison, Parallel (advanced) costs $0.084 per task and Brave (LLM context) costs $0.13 per task
AGI will be here when agents learn to multitask across >10 tasks at once without steering. Today's models seem to often operate on a last in first out stack and lose track of the long-term vision, hampering their ability to plan across long engineering tasks. Asking agents to take notes helps somewhat but also seems like an incomplete solution.
Back when I was working in finance, it took weeks to acquire and test one new data source. Now financial research, modeling, backtesting, etc. is now infinitely easier and faster with AI. Creativity is the bottleneck rather than code.
We are doubling down on making Computer more useful for financial researchers and analysts.
Computer now connects to new licensed data sources, including Dun & Bradstreet, Guidepoint, IBISWorld, and 20+ others. An analyst can ask Computer a question and get an answer built from the firm's own licensed sources without APIs or separate logins to work through.
Every figure traces to the source record it came from. Teams doing high-stakes work would know exactly which data set produced which number.
To connect, select a provider in Settings → Connectors and sign in with an existing license. Available now to all Perplexity users.
perplexity.ai/hub/blog/compu…
Congrats to everyone involved from both Nvidia and Perplexity. Excited to build great products that leverage both local and cloud compute simultaneously. Much more to come!
Today we’re launching Portable Computer on @NVIDIA DGX Spark.
Portable Computer is a fully local version of Perplexity Computer, where the entire runtime: orchestrator LLM, subagent LLM, agent harness all run on your local hardware. No cloud dependency.
Glad that the ecosystem has converged on an open standard for plugins! Incredibly easy to install or build your own.
The @perplexity_ai API MCP server is now available as an Agent Plugin 🎉
Install it in Codex, Cursor, VS Code, and other Agent Plugins 1.0.0 conformant clients. Your agent gets real-time web search, reasoning, and research through Perplexity's API.
Visit the repo to get started: pplx.ai/api_agent_plugin
If your agent needs frontier web search, it's hard to beat Perplexity's Agent API (on any metric).
Introducing Web Search Benchmarks 🌐
Rankings of search tools across different models and configurations to help you decide how to ground your agent: openrouter.ai/benchmarks
Open weights continues to nearly match the frontier while edging the frontier out on several benchmarks.
Time to try GLM out for coding!
Introducing GLM-5.3: Built to Code. Ready for Cyber Defense.
- Top-tier coding and agentic capabilities, achieved through post-training on the 743B base model
- A major leap in cybersecurity, setting a new standard among open models
Tech Blog: z.ai/blog/glm-5.3
How and why are now the hardest parts of engineering: how to measure quality, how to design systems, how to find levers to pull, and why it all matters.
blog.senko.net/code-was-neve…
