frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Do you ever ask "Please Claude I need this my account is kinda tokenless "

https://engram-three.vercel.app/
1•-Refraction-•1h ago

Comments

-Refraction-•1h ago
Literally the title. Ever sat in front of your computer screen, watching Claude burn tokens reading shit it has already read over and over again? Burning tokens that could have been used to write code or fix bugs?

That, my good friend, is exactly what Engram fixes. Engram is a shared context intelligence system that lets you reduce your token burn rate by a factor of huge magnitudes.

We measured this on a real codebase (42 TypeScript files, ranging from 1 KB to 42 KB per file).

For a task like "fix the billing page," an agent without Engram reads roughly 5 large files to orient itself, the kind that run 6,000–7,000 tokens each. That's around 31,500 tokens before writing a single line. With Engram, get_context_for_task returns the ranked list of relevant symbols across all 42 files in about 2,700 tokens. The agent then reads the one file it actually needs around 2,800 tokens for a large component. Total: around 5,500 tokens. That's roughly 6× cheaper, and the agent goes straight to the right file instead of guessing.

For individual symbol lookups, get_symbol returns a function's signature, callers, callees, and test associations in about 95 tokens. A small utility file in this codebase costs around 450 tokens to read in full, whereas a large component is 6,000+. If an agent needs 10 symbol lookups across 10 small files, that's 950 tokens with Engram vs 4,500 without. About 4× cheaper at the low end, and the savings compound sharply as files get larger.

For a targeted task like "fix the querySymbol function," the agent normally reads 4 core files, which adds up to around 26,000 tokens. With Engram, the agent makes 3 tool calls returning just the exact definitions, callers, and blast radius, for about 320 tokens. That's roughly 80× cheaper for the same task-level understanding.

Every. Single. Session.

It even knows what changed since your last session so after a git pull, instead of rereading everything again, your agent gets a delta: "these 14 symbols changed, here's what's new." Cold starts become warm starts.

It even shows you the blast radius of a change before you make it.

get_dependencies("UserService", depth: 3) tells you exactly what breaks. No more surprise regressions at runtime.

It even works across agents. Claude Code, Cursor, Windsurf, Cline, Zed, Gemini CLI. One MCP server, everything speaks to it.

The free plan is now officially ready, no credit card required. It runs 100% locally on your code which never leaves your machine. The paid tiers are for cloud sync and team graph sharing, which you only need when you have teammates and want to share and sync codebase context across agents across teams, and across computers.

go to the website and sign up for your free account now and stop wasting those tokens!

That's it.

OpenClaw vs. Hermes Agent: The race to build AI assistants that never forget

https://thenewstack.io/persistent-ai-agents-compared/
1•thunderbong•4m ago•0 comments

Palimpsest

https://en.wikipedia.org/wiki/Palimpsest
1•tosh•4m ago•0 comments

A catastrophic climate event is upon us. Here is why you've heard so little

https://www.theguardian.com/commentisfree/2026/apr/23/catastrophic-climate-event-scientists-atlan...
1•Michelangelo11•5m ago•0 comments

Residual Malicious Payloads on HackerOne After Vulnerability Fixes

https://hackerone.com/reports/3168691
1•nathan84•11m ago•0 comments

Show HN: NoonFlow – a macOS workspace I built for Claude Code and Codex

https://github.com/heyallencao/NoonFlow/releases
1•AllenCao•12m ago•3 comments

Political Discontent in China Is Associated with Isolating Personality Traits

https://www.journals.uchicago.edu/doi/abs/10.1086/719273
1•xnhbx•15m ago•0 comments

Apple's New CEO Must Launch 5 Product Ecosystems or Miss the Next iPhone Moment

https://greentic-ai.medium.com/apples-next-ceo-must-launch-these-5-product-ecosystems-or-miss-the...
1•mectors•16m ago•0 comments

Show HN: Playing music over movies causes them to sync

1•e1ghtSpace•16m ago•0 comments

Colorado citizens pay with rights, xAI pays nothing, long live democracy

https://aitwerp.com/signals/xai-colorado-discrimination-federal/
1•Inziu•21m ago•1 comments

Hire based on the conversation about code, not the code itself

https://dbarabashh.com/thoughts-and-experience/hire-for-the-conversation-not-the-code
1•birdculture•27m ago•0 comments

LogAct: Enabling agentic reliability via shared logs

https://arxiv.org/abs/2604.07988
1•pramodbiligiri•34m ago•0 comments

Show HN: A faithful offline recreation of the classic MS-DOS Editor

https://pascar.run
1•mysticmode•52m ago•0 comments

Managing the Unmanaged Switch

https://watchmysys.com/blog/2026/03/managing-the-unmanaged-switch/
2•luu•53m ago•0 comments

Show HN: I fixed sleep schedule by tracking solar time, so I built an app for it

https://sun.mikolajmocek.com/
1•mondonno•55m ago•0 comments

Richard Sutton – Father of RL thinks LLMs are a dead end [video]

https://www.youtube.com/watch?v=21EYKqUsPfg
3•mpweiher•55m ago•1 comments

Scaling Camera File Processing at Netflix

https://netflixtechblog.com/scaling-camera-file-processing-at-netflix-6dab2b1e80be
2•redblueflame•59m ago•0 comments

Show HN: Agent MCP Studio – build multi-agent MCP systems in a browser tab

https://www.agentmcp.studio
3•stealthtsdb•59m ago•0 comments

Testing GPT-5.5 in early access: what we are seeing so far

https://lovable.dev/blog/gpt-5-5-now-in-lovable
2•doener•1h ago•1 comments

Escrow Security for iCloud Keychain

https://support.apple.com/guide/security/escrow-security-for-icloud-keychain-sec3e341e75d/web
2•gurjeet•1h ago•0 comments

Tewart Brand on LSD, A.I. Black Boxes and the Beauty of Care [video]

https://www.youtube.com/watch?v=t8u24wvHeSE
1•born-jre•1h ago•0 comments

Code review advice for vibe coders

https://xata.io/blog/code-review-for-vibe-coders
2•tee-es-gee•1h ago•0 comments

Show HN: A CLI to use any model in your coding agent

https://getaivo.dev/
2•spirit23•1h ago•0 comments

What is Nostr? A simple guide to the protocol

https://usenostr.org/
2•vlugorilla•1h ago•0 comments

The Tiny Donut That Proved We Still Don't Understand Magnetism [video]

https://www.youtube.com/watch?v=XKSjCOKDtpk
2•mpweiher•1h ago•0 comments

List of personal sites that host Wander console, a tool to explore the small web

https://susam.codeberg.page/wander/wcn.html
2•susam•1h ago•0 comments

Naming Things Is Easy Now

https://notesbylex.com/naming-things-is-easy-now
3•lexandstuff•1h ago•0 comments

I left Vercel Pro ($20/mo) for a $10/mo VPS. 7-day Next.js migration report

https://gist.github.com/Samarth0211/b728534af45242b61b45a87a4ecdf155
1•samarth0211•1h ago•1 comments

Global Energy Flows

https://ig.ft.com/global-energy-flows/
1•saswatms•1h ago•0 comments

Mystery Cpuid Bit

http://www.os2museum.com/wp/mystery-cpuid-bit/
1•userbinator•1h ago•0 comments

Do you ever ask "Please Claude I need this my account is kinda tokenless "

https://engram-three.vercel.app/
1•-Refraction-•1h ago•1 comments