# What's hyped now, what already peaked, and which company you're in (slim) > Vlad Cealicu on what people who work with agents talk about in November 2026 and what they've stopped talking about, kept to what he could back up and sorted by who's talking. Harness engineering, cost per solved task, two to four parallel coding agents, weight-space programming and open-weight models at the frontier. Forward deployed engineers, AI code review, chief-of-staff agents and desktop agents at the companies that embraced AI. Prompt engineering jobs, which model is best, token leaderboards and stacking MCP servers at the ones that are behind. With decision models, 25 of them in eight weeks, as the measure of how fast it moves. - Full: https://www.anchorterminal.com/blog/what-is-hyped-now.md (~4,350 tokens) · this version ~1,780 tokens · JSON https://www.anchorterminal.com/blog/what-is-hyped-now.json · canonical https://www.anchorterminal.com/blog/what-is-hyped-now - Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · MCP: https://www.anchorterminal.com/mcp · Updated: 2026-11-13 - Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data) By Vlad Cealicu*, 13 November 2026. 1,597 words, about 7 min to read. - 15 September: when TypeSafe's Jev shipped, the first of the decision models we list - 25: decision models we list that shipped in the eight weeks from that day - 2 to 4: coding agents most of Kilo's engineers run in the foreground, against the 50 to 100 people post about - 35%: of teams in Flux's survey who write code with AI and won't ship it Decision models, parallel agents and AI code. The wave of decision models started with TypeSafe's Jev on 15 September 2026, and eight weeks later we list 25 of them, with OpenAI's on 6 October, Microsoft's on 9 October and Google's on 11 November [[1]](#ref-1). I started with a list of 24 things, 12 that I thought were hyped and 12 that I thought had already peaked, and when I went looking for something to back each one, 10 didn't hold up, so they're not here. ## What was being argued about in May On 28 May Fortune called tokenmaxxing over, after a spring in which big companies ranked engineers by tokens spent, Meta's internal leaderboard was reported in April and came down two days later, and Uber got through its whole 2026 token budget in four months [[2]](#ref-2). ## The frontier and the AI-native teams Start with harness engineering. The cost they count is the cost of a finished task. Coding agents run in parallel, and fewer of them than the posts say. The newest thing they talk about is weight-space programming. Open-weight models are on this side too. ## The companies that embraced AI Forward deployed engineers are what these companies are hiring. They also have a review problem, because the writing worked. Chief-of-staff agents are the other thing they're buying. ## The companies that are behind If the conversation at your company is one of these, I think it's behind. Connecting every MCP server you can find went earlier than that. ## How fast it's moving When each thing was named, measured or called over 1. **6 May 2025** (on the way out). Fast Company says prompt engineering as a job has all but disappeared [[16]](#ref-16) 2. **20 February 2026** (on the way out). Cloudflare's Code Mode, the second fix for stacking MCP servers [[4]](#ref-4) 3. **28 May 2026** (on the way out). Fortune calls tokenmaxxing over [[2]](#ref-2) 4. **30 July 2026** (on the way in). TechCrunch on the rush for forward deployed engineers [[11]](#ref-11) 5. **13 August 2026** (on the way in). DoiT's cost per task framework [[6]](#ref-6) 6. **15 September 2026** (on the way in). Jev, the first of the decision models [[1]](#ref-1) 7. **2 November 2026** (on the way in). Weight-space programming gets its name [[9]](#ref-9) 8. **11 November 2026** (on the way in). Google's decision model, the 25th since Jev [[1]](#ref-1) Prompt engineering took about two years to go from the hot new job to all but gone [[16]](#ref-16), the token leaderboards were news in April and called over by the end of May [[2]](#ref-2), and decision models are eight weeks old with 25 on our list [[1]](#ref-1). Decision models we list, by the month they first shipped | Month | Models | | --- | --- | | Before September 2026 | 2 | | September 2026 | 8 | | October 2026 | 16 | | November 2026, to the 13th | 1 | Each series has a zero-based scale. The two from before September are an IBM model from May and a classifier from 2024 that do the same job and sit in the same category [[1]](#ref-1). What's missing here is our own numbers for the rest. If you want to know which company you're in, write down the last three things your team argued about on AI and see which section they fall in. --- More in [Opinion](https://www.anchorterminal.com/blog/opinion/index.md) · Older opinion [Qwen4-Max is in preview, and Alibaba's own docs still stop at Qwen3.8](https://www.anchorterminal.com/blog/qwen4-max-preview-docs-still-say-3-8.md) - Older: [How we built Agent of Somebody](https://www.anchorterminal.com/blog/how-we-built-agent-of-somebody.md) - Newer: [I don't think you need Claude's harness](https://www.anchorterminal.com/blog/another-harness-for-business.md) - All posts: https://www.anchorterminal.com/blog/index.md Anchor Terminal grades the tools AI agents run on, with a source for every score. ## References Read on 13 November 2026. Quotes are as published. [1] Anchor Terminal, decision model listings, each with its first release date and the vendor pages it was graded from, counted on 13 November 2026. Source for Jev on 15 September, the 25 that shipped from then to 11 November, the dates for OpenAI, Microsoft and Google, the count by month, and the two older models in the category. [2] Anchor Terminal, A machine that runs all night, and why the labs aren't at the forefront, 7 November 2026, with Fortune's reports of 9 April and 28 May 2026, LeadDev's of 17 August 2026 and Zhang and others (arXiv 2605.23950) as its sources. [3] Kilo Code blog, Darko Gjorgjievski, 26 May 2026. Are parallel coding agents actually useful, or just Twitter theater? Source for the posts claiming 50 to 100 agents, the seven senior engineers asked, the two to four foreground agents most of them run, and Kilo's conclusion. [4] MCP.Directory, 4 May 2026. MCP context bloat, the 2026 fixes that work. [5] He, C., Zhou, X., Wang, D., Xu, H., Liu, W. [6] DoiT, 13 August 2026. Cost per task, not cost per token. [7] Anchor Terminal, how our listing research runs, November 2026, not published. [8] Anthropic, claude.dev blog, 9 November 2026. Safely scaling multi-agent teams with Claude Fable 5.2. Source for the 1,000-agent run being described as a toy experiment. [9] Takehi, R., Teoh, J., Zhang, A. [10] Anchor Terminal, open-weight model listings, each with its first release date, counted on 13 November 2026, the first of them listed on 20 October 2026. Source for the count of listed models that first shipped in July, August, September and October 2026. https://www.anchorterminal.com/categories/open-weights [11] TechCrunch, Rebecca Bellan, 30 July 2026. Forward-deployed engineers are the AI industry's latest talent obsession. [12] Built In, Lauren Hanford, 2026. 96 percent of developers don't trust AI-generated code. [13] LeadDev, Chantal Kapani, 30 June 2026. AI-generated code sparks production confidence crisis. [14] The New Stack, 4 June 2026. Asana says its new AI chief of staff turns your Slack chaos into trackable work. [15] Google, Gemini API changelog. [16] Fast Company, Henry Chandonnet, 6 May 2025. On prompt engineering quickly going extinct. [17] Anchor Terminal, the Railway MCP server listing. [18] Anchor Terminal listings data, counted on 13 November 2026, not published as a table.