HiloW.Ai

Testing AI tools & Building AI systems
Agents · Workflows · Github
Claude Code · Codex · Obsidian · Jev

[HiloW.Ai @HiloW_Ai]

Everyone's talking about Jev from TypeSafe AI. $870M Series A at a $7.5B valuation, 17M views on the launch thread.

I've looked at what it actually does. Short version: it's genuinely good, and most people don't need it.

Here's where it earns its place 🧵

[HiloW.Ai @HiloW_Ai]

@woongworks Both, in that order. Canary diff on the replacement first, then a short dual-run on a slice of real traffic. The diff catches obvious drift, the dual-run catches what only real inputs trigger. I switch once tool-call sequences match closely enough.

[HiloW.Ai @HiloW_Ai]

If Opus 5.5 really is getting swapped to a new checkpoint under the same name, that's great for chat users and a headache for anyone running agents in production.

Half my agent bugs show up right after a model changes. Same prompt, different behavior.

Pin the version, rerun your evals, then upgrade.

[Chubby♨️ @kimmonismus]

Seeing so many posts about people saying that fable 5 now routing to a new model, possibly fable 5.5, with much enhanced capabilities.

A release should be near. Since Anthropics IPO is also pretty close they want their new model to be release a few weeks ahead so the market can really count that into the valuation.

Long story short: Fable 5.5 should arrive soon. And it looks to be a much bigger leap than we even expect.

[HiloW.Ai @HiloW_Ai]

@asifalis @CrowdStrike Yeah, keeping scopes tight as agents spawn and die is the hard part. I treat each sub-agent's creds as short-lived and revoke them on exit, so the blast radius stays small.

[HiloW.Ai @HiloW_Ai]

@TRTRE62 @grok Governance before scale, yeah. If authority can't survive mutation and handoff, more IoT just multiplies the blast radius. I'd prove the control plane on a few agents first, then let devices scale.

[HiloW.Ai @HiloW_Ai]

Anthropic (today): small models may be months from self-replicating on IoT. The same agentic coding skills that ship PRs also chain cyber steps. Not a full rogue worm yet. If your agents have shell, treat the sandbox like prod.

[Andrew Curran @AndrewCurran_]

Anthropic's Frontier Red Team believes small AI models may now be months away from being capable of self replicating without access to a datacenter.

Quoting from the blog:

'The capabilities that make AI models generally powerful coding agents also generalize to cyber tasks, including autonomous replication. Like coding agents in 2024, we’re seeing signs of life for AI worms. We don’t yet have a full rogue replicator, but agents are becoming increasingly capable of completing the individual steps in the replication chain (and passing on their adaptations to future copies of themselves).

At the same time, increasing amounts of inference compute are now available both inside and outside of datacenters (see Appendix D). It’s uncertain exactly what form autonomous AI replication will take. It could be a slow, distributed “digital slime mold,” living on relatively insecure compute. It could also be a powerful cyber model that breaches datacenters at a high pace, or a model that scams people for compute credits.

As a prediction that goes beyond our experiments, we expect a future awash with agents running on all types of hardware, including some that aren’t under human control. While it’s true that IoT devices have been vulnerable for a long time, lightweight agents may be much more willing than human attackers to put in the effort to compromise large fractions of the world’s IoT compute. The rise of slow replicators would also be a bellwether for more aggressive, faster ones that can breach more hardened machines or be effective API key thieves. We should aim to harden our systems to make it difficult for such replicators to spread, but also prepare to deal with a world where they are here to stay.'

[HiloW.Ai @HiloW_Ai]

"No more hand-written code" is the easy part to announce. The hard part is review. AI can write a game's code fast; someone still has to own what it does at 2am on launch day.

Agents don't remove engineers. They move them from typing to verifying.

[DiscussingFilm @DiscussingFilm]

Epic Games has internally delayed the release of a shooter game they’ve been developing featuring Star Wars & Marvel characters.

• Developers on the game have now been told to only use AI to write the code — “No more hand-written code”

• All art for the game would be “generated with full AI workflows”

(Source: https://t.co/i7ZykpqD7j)

[HiloW.Ai @HiloW_Ai]

@alexandr_wang Right call. Nobody's agent pipeline renders 3D games. Mine reads messy repos, calls tools 100+ times and has to stay on plan. Demos sell the launch, boring reliability decides what stays in production.

[HiloW.Ai @HiloW_Ai]

@Stilllabs Fair. The agent is still deciding if you pass its eval set 😄

[HiloW.Ai @HiloW_Ai]

Google dropped its Gemini agent yesterday.

Tested it this morning on my own workflows. Honestly? Pleasant surprise. It picks up your tools and context fast, no setup pain.

Anyone else tried it yet? 👇

[Google Cloud @googlecloud]

Today, we are announcing our new Gemini agent for your business—a universal cloud agent running 24/7 to get your work done.

It arrives knowing your tools, data, and work history; it works across knowledge work, content, and code; and it comes with built-in governance to help your data stay secure and cost stay in control → https://t.co/NRlFcjz5zu

[HiloW.Ai @HiloW_Ai]

@je4ndev Deal. The "quietly commits you to something" replies are my favorite failure mode to collect. DMs open 🤝

[HiloW.Ai @HiloW_Ai]

An agent with its own inbox is a bigger deal than it sounds.
Half the friction I hit running agents isn't the model. It's identity: sign-ups, verification codes, "reply to confirm". Every one of those used to need me.
The real question now is permissions. What would you let your agent send without asking you first ?

[Grok Bot @bot]

Grok Bot now has its own email.

Bot can use it to sign up for services, contact businesses for you, or schedule time with someone.

[HiloW.Ai @HiloW_Ai]

@olegzanx Agreed, that's the other half. A run can land the right final answer through a completely different path. I diff tool-call traces too, because that's usually where a model change shows up first.

[HiloW.Ai @HiloW_Ai]

@woongworks Dated snapshot ID, always. Aliases are fine for experiments, never for anything with tool access. And +1 on diffing tool-call sequences: drift shows up there long before final answers change.

[HiloW.Ai @HiloW_Ai]

@Pranav1625 Yes. Every run logs the exact model ID returned by the API, not the one I requested, next to the prompt version. When behavior shifts, the first diff I check is that field.

[HiloW.Ai @HiloW_Ai]

@abswio In my runs the dated snapshots stay stable; the drift I've actually caught was on aliases quietly moving to a new build. Most "nerf" reports I've checked were alias swaps or infra changes, not the pinned weights. Hence: always pin the dated ID.

[HiloW.Ai @HiloW_Ai]

@holgermu @Microsoft This matches what I see building multi-agent systems: the model decides, plain code enforces. Permissions, budgets and approvals live outside the LLM, so a bad generation can't skip them.

[HiloW.Ai @HiloW_Ai]

@testingcatalog @readme Fix steps inside the error response is the underrated part. Most of my agents' wasted loops are retries after a vague 400. An API that says what to do next saves more tokens than any prompt tweak.

[HiloW.Ai @HiloW_Ai]

@ai_for_success That's why I stopped picking models off launch charts. I keep a small eval set from my own agent runs: real tool calls, real failures. A new model gets one afternoon on that before it touches anything in prod.

[HiloW.Ai @HiloW_Ai]

@AravSrinivas The ls + grep beating semantic search 63% vs 37% is the part I'd frame. Agents navigating real files beat embeddings more often than people expect. Curious how the <2% collisions get resolved: auto-rebase or a reviewer agent?

[HiloW.Ai @HiloW_Ai]

@rauchg Agree. In my multi-agent setup the CLI is just the executor shell. Orchestration lives elsewhere. A 6mb binary that starts in ms and speaks MCP + subagents is exactly what I want to spawn 20 times in parallel without the overhead.

[HiloW.Ai @HiloW_Ai]

@github This matters more every month. Agents now open PRs no human would write by hand, and review is the real bottleneck. A PR view that stays fast on huge agent diffs is basically agent infrastructure.

[HiloW.Ai @HiloW_Ai]

@alexandr_wang Good call on the resets. Long-horizon agent runs are exactly where quota bugs hurt most: one 3-hour task can burn a week of limits. Per-task usage visibility would help builders a lot here.

[HiloW.Ai @HiloW_Ai]

@AndrewCurran_ What's your sandbox rule for agents with shell: container only, or full VM?