frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Show HN: Spinf-serve – inference server for bulk scoring and decisions

https://github.com/spinfinc/spinf-serve
1•giarec•1m ago•0 comments

Show HN: WireWhy – a browser wiring simulator that explains the "why"

https://wirewhy.com/
1•hindus•2m ago•0 comments

Takto One – open-source Hand Exoskeleton

https://github.com/molanocortes/takto-one/tree/main
1•_Microft•3m ago•0 comments

Ask HN: What's it like working at Google nowadays? Core teams like Search, Ads?

1•karakoram•4m ago•0 comments

Bring your website or project to Lovable

https://lovable.dev/free-import
1•doener•5m ago•0 comments

What YC's Minecraft and Mindustry agent swarms show about multi-agent teams

https://theainewsreport.com/2026-10-30-yc-agent-swarms-minecraft-mindustry-explained.html
1•titusblair•5m ago•0 comments

Disk I/O profiling with eBPF

https://coroot.com/blog/disk-io-profiling-with-ebpf/
2•nikolay_sivko•7m ago•0 comments

Binance hot wallet signed 50 EIP-7702 delegations 3 minutes before the drain

https://www.reddit.com/r/CryptoCurrency/comments/1xj0mp0/binance_hot_wallet_signed_50_eip7702_del...
1•Onavo•7m ago•0 comments

Show HN: Teaching a 4M-param model to read like a child, with RL and no labels

https://rjb7731.github.io/blog/2026/learning-from-scratch/
1•rybthrow2•8m ago•0 comments

Check for IOCs for the ongoing Claude Code extensions exfil campaign

https://github.com/cookiengineer/check-claude-code-exfil
1•cookiengineer•8m ago•0 comments

Recent and Cumulative Cannabis Use and Sleep in Middle-Aged Adults

https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2855118
1•bookofjoe•11m ago•0 comments

More thinking isn't (always) better

https://www.ariwilson.com/writing/agent-effort-levels/
1•ariwilson•12m ago•1 comments

Signet: Cryptographic Enforcement of LLM Transitions

https://github.com/harishbalasubramanian/signet-ai
1•sterling-vance•13m ago•0 comments

What happens when cloud coding agents are as fast as your local machine?

https://opencomputer.dev/blog/cloud-agents-as-fast-as-local/
1•iacguy•13m ago•0 comments

The $200M Machine that Prints Microchips [video]

https://www.youtube.com/watch?v=B2482h_TNwg
1•tosh•15m ago•0 comments

U.S. Oil Industry Earned over $130B in Profit During Iran War

https://www.nytimes.com/2026/10/30/business/energy-environment/exxon-chevron-oil-profits-iran-war...
7•ceejayoz•16m ago•0 comments

First video gen model trained on real neurons online

https://twitter.com/AlexKsendzovsky/status/2116153410101944352
1•philippb•17m ago•0 comments

Show HN: I built a safety gate for autonomous coding agents

https://inbin-gate.vercel.app/
1•emmanueltsakpo•18m ago•0 comments

Show HN: Hashsync – Deduplicate SSH streams against the receiver's files

https://github.com/someplaceguy/hashsync
1•someplaceguy•19m ago•0 comments

OpenAI Private Intelligence

https://openai.com/business/private-intelligence/
1•pretext•21m ago•0 comments

A Working Nintendo in ~1GB of Qwen Weights

https://github.com/dbreunig/nesformer
10•dbreunig•24m ago•5 comments

Voyager 1 gives controllers the silent treatment

https://www.theregister.com/science/2026/10/30/voyager-1-gives-controllers-the-silent-treatment/5...
4•ohjeez•25m ago•0 comments

Show HN: NKTg AI – An LLM that decodes text before generating

https://nktg.org/
1•NKTgAI•25m ago•0 comments

Google Jules Turndown on December 31, 2026

https://jules.google/docs/changelog/2026-10-29
3•simonpure•26m ago•0 comments

A geography game where you guess cities based on compass arrows

https://visitwhale.com/city-angle/
1•whalegames•26m ago•0 comments

Package science into reusable, citable knowledge

https://www.aluna.ai
1•Bluestein•28m ago•0 comments

Product Hunt is swamped now

https://www.mahl.me/blog/product-hunt-is-swamped-now/
1•monneyboi•29m ago•0 comments

Interface Lineage – library of interface patterns and principles

https://interfacelineage.com/
1•ChrisArchitect•29m ago•0 comments

Show HN: Shell-pet – a hermit crab for your README that reacts to repo activity

https://shell-pet.vercel.app
1•amhtutt•29m ago•0 comments

OpenAI CEO Sam Altman Says You Having DRAM Instead of Data Centers Is "Terrible"

https://www.thegamer.com/openai-ceo-sam-altman-consumers-having-dram-terrible/
9•josephcsible•29m ago•3 comments
Open in hackernews

Ask HN: What is your best hack to keep your agent's context in sync?

1•heysrb•1h ago
A lot of people manually maintain markdown files and use them to sync context. Some people even use Obsidian.

I was wondering what is the best hack you have discovered to keep the agents in sync? And how does it perform, does it work reliably well?

One of the problems I face is multiple agents writing to a common memory which can lead to unwanted memory overwrites without any trace.

Comments

bbor•43m ago
I was gonna wax poetic, but I think this explains it pretty well -- the bulletin and the campaign dirs get more usage than the rest of directory combined, I think. I've completely disabled all harness-internal memory mechanisms, as they're too opaque, make weird assumptions about TTL for random one-off comments, and (obv) aren't cross-harness.

  ~/ai/...
      campaigns
      dispatch
      evidence
      reports
      scratch
      telemetry
      worktrees
      quarantine
      BULLETIN.md
      README.md