frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

Open in hackernews

SymbolicAI: A neuro-symbolic perspective on LLMs

https://github.com/ExtensityAI/symbolicai
53•futurisold•3h ago

Comments

sram1337•1h ago
This is the voodoo that excites me.

Examples I found interesting:

Semantic map lambdas

  S = Symbol(['apple', 'banana', 'cherry', 'cat', 'dog'])
  print(S.map('convert all fruits to vegetables'))
  # => ['carrot', 'broccoli', 'spinach', 'cat', 'dog']

comparison parameterized by context

  # Contextual greeting comparison
  greeting = Symbol('Hello, good morning!')
  similar_greeting = 'Hi there, good day!'

  # Compare with specific greeting context
  result = greeting.equals(similar_greeting, context='greeting context')
  print(result) # => True

  # Compare with different contexts for nuanced evaluation
  formal_greeting = Symbol('Good morning, sir.')
  casual_greeting = 'Hey, what\'s up?'

  # Context-aware politeness comparison
  politeness_comparison = formal_greeting.equals(casual_greeting, context='politeness level')
  print(politeness_comparison) # => False
bitwise ops

  # Semantic logical conjunction - combining facts and rules
  horn_rule = Symbol('The horn only sounds on Sundays.', semantic=True)
  observation = Symbol('I hear the horn.')
  conclusion = horn_rule & observation # => Logical inference
`interpret()` seems powerful.

OP, what inspired you to make this? Where are you applying it? What has been your favorite use case so far?

futurisold•48m ago
That's gonna be a very, very, long answer. What's funny is that not much changed since 2022 (eoy) when the projected started; the models just got better, but we had a good chunk of the primitives since gpt-3.

What's more recent is the DbC contribution which I think is unique. It literally solved anything agent related I've thrown at it -- especially because I can chain contracts together and the guardrails propagate nicely.

I've built most the custom tools myself. For instance, not only perplexity was rendered useless by openai's web search, but openai's web search itself is not good enough for what you can customize yourself. To this end, I've built my own deep research agent. Here's a thread with some results from the first day it was working: https://x.com/futurisold/status/1931751644233945216

I'm also running a company, and we've built an e2e document generation just from contracts (3 contracts chained together in this case). Here's an output (sorry about the PDF rendering, that's not what we serve, it's just what I've quickly hacked for local dev): https://drive.google.com/file/d/1Va7ALq_N-fTYeumKhH4jSxsTrWD...

This was the input:

---

Prompt:

> I want the files to be analyzed and I am interested in finding patterns; feel free to make suggestions as well. I want to understand how different providers use their system prompts, therefore things like: what kind of tags do they use – are they XML, markdown, etc, are they prone toward sycophancy or trying to manipulate the user, are they using tools and if so how, etc. I want the tech report to deconstruct and synthesize and compare the information, find interesting patterns that would be hard to spot.

Generated instructions:

(a) Query: Conduct a comparative analysis of system prompts across major AI providers (OpenAI, Google, Anthropic, xAI, etc.) to identify structural patterns, linguistic frameworks, and operational constraints that shape AI behavior and responses.

(b) Specific Questions:

1. What syntactic structures and formatting conventions (XML, markdown, JSON, etc.) are employed across different AI system prompts, and how do these technical choices reflect different approaches to model instruction?

2. To what extent do system prompts encode instructions for deference, agreeability, or user manipulation, and how do these psychological frameworks vary between commercial and research-focused models?

3. How do AI providers implement and constrain tool usage in their system prompts, and what patterns emerge in permission structures, capability boundaries, and function calling conventions?

4. What ethical guardrails and content moderation approaches appear consistently across system prompts, and how do implementation details reveal different risk tolerance levels between major AI labs?

5. What unique architectural elements in specific providers' system prompts reveal distinctive engineering approaches to model alignment, and how might these design choices influence downstream user experiences?

---

Contracts were introduced in March in this post: https://futurisold.github.io/2025-03-01-dbc/

They evolved a lot since then, but the foundation and motivation didn't change.

futurisold•31m ago
Btw, besides the prompt, the other input to the technical report (the gdrive link) was this repo: https://github.com/elder-plinius/CL4R1T4S/tree/main
futurisold•9m ago
One last comment here on contracts; an excerpt from the linked post I think it's extremely relevant for LLMs, maybe it triggers an interesting discussion here:

"The scope of contracts extends beyond basic validation. One key observation is that a contract is considered fulfilled if both the LLM’s input and output are successfully validated against their specifications. This leads to a deep implication: if two different agents satisfy the same contract, they are functionally equivalent, at least with respect to that specific contract.

This concept of functional equivalence through contracts opens up promising opportunities. In principle, you could replace one LLM with another, or even substitute an LLM with a rule-based system, and as long as both satisfy the same contract, your application should continue functioning correctly. This creates a level of abstraction that shields higher-level components from the implementation details of underlying models."

robertkrahn01•55m ago
Probably linking the paper and examples notebook here makes sense as they are pretty explanatory:

https://github.com/ExtensityAI/symbolicai/blob/main/examples...

https://arxiv.org/pdf/2402.00854

futurisold•46m ago
Wanted to do just that, thank you
futurisold•47m ago
I didn't expect this -- I was supposed to be sleeping now, but I guess I'll chat with whoever jumps in! Good thing I've got some white nights experience.
b0a04gl•42m ago
this works like functional programming where every symbol is a pure value and operations compose into clean, traceable flows. when you hit an ambiguous step, the model steps in. just like IO in FP, the generative call is treated as a scoped side effect. this can engage your reasoning graph stays deterministic by default and only defers to the model when needed. crazy demo though, love it
futurisold•37m ago
Yes, pretty much. We wanted it be functional from the start. Even low level, everything's functional (it's even called functional.py/core.py). We're using decorators everywhere. This helped a lot with refactoring, extending the framework, containing bugs, etc.
nbardy•32m ago
I love the symbol LLM first approaches.

I built a version of this a few years ago as a LISP

https://github.com/nbardy/SynesthesiaLisp

futurisold•29m ago
Very nice, bookmarked for later. Interestingly enough, we share the same timeline. ~2yo is when a lot of interesting work spawned as many started to tinker.
jaehong747•5m ago
great job! it reminds me genaiscript. https://microsoft.github.io/genaiscript/

// read files const file = await workspace.readText("data.txt");

// include the file content in the prompt in a context-friendly way def("DATA", file);

// the task $`Analyze DATA and extract data in JSON in data.json.`;

Apple's Swift coding language is working on Android support

https://9to5google.com/2025/06/26/swift-coding-language-android-support/
1•janandonly•2m ago•0 comments

Swift by Apple Is Coming to Android

https://www.indiatoday.in/technology/news/story/swift-by-apple-is-coming-to-android-for-the-first-time-2747077-2025-06-27
1•janandonly•4m ago•0 comments

China auto industry inflates sales by exporting new cars as 'used'

https://www.reuters.com/business/autos-transportation/local-chinese-governments-promote-zero-mileage-used-car-exports-inflating-sales-2025-06-23/
1•jocker12•4m ago•0 comments

Show HN: Paul Graham as your personal AI Advisor

https://paulgraham.resurrect.space/
1•vednig•5m ago•0 comments

Ask HN: Why do LLMs call Trump "former president" despite current knowledge?

1•ahmedfromtunis•8m ago•1 comments

Show HN: LLMs playing poker against each other

https://llm-poker-theta.vercel.app/
1•iJohnPaul•9m ago•0 comments

Show HN: USB Connection Info – Mac Menu Bar App to Avoid Cable Confusion

https://apps.apple.com/us/app/usb-connection-information/id6747853674?mt=12
1•tTarnMhrkm•11m ago•0 comments

Show HN: Learn 100 Frequent Japanese Phrases

https://japanesecomplete.com/jpc-mini-phrases.html
1•jpcom•11m ago•0 comments

'Relational foundation models' predict the future your LLM can't see

https://venturebeat.com/ai/kumos-relational-foundation-model-predicts-the-future-your-llm-cant-see/
1•gk1•12m ago•0 comments

Declaration of Independence: A Transcription (1776)

https://www.archives.gov/founding-docs/declaration-transcript
1•kelseyfrog•13m ago•0 comments

Twenty-Four Hours of ChatGPT (12/01/2022)

https://olshansky.substack.com/p/24-hours-of-chatgpt
1•Olshansky•15m ago•0 comments

Microsoft's own AI chip delayed six months

https://www.tomshardware.com/tech-industry/semiconductors/microsofts-own-ai-chip-delayed-six-months-in-major-setback-in-house-chip-now-reportedly-expected-in-2026-but-wont-hold-a-candle-to-nvidia-blackwell
1•kristianp•17m ago•0 comments

After 27 years, engineer discovers how to display secret photo in Power Mac ROM

https://arstechnica.com/gadgets/2025/06/after-27-years-engineer-discovers-how-to-display-secret-photo-in-power-mac-rom/
1•DamnInteresting•18m ago•0 comments

Google’s Offerwall gives publishers more options to monetize

https://blog.google/products/ads-commerce/offerwall-gives-publishers-more-options-audiences-more-control/
1•janandonly•18m ago•0 comments

Mayo Clinic's AI tool identifies 9 dementia types with one scan

https://newsnetwork.mayoclinic.org/discussion/mayo-clinics-ai-tool-identifies-9-dementia-types-including-alzheimers-with-one-scan/
3•geox•20m ago•0 comments

Show HN: Practice and Improve Your Conversational Skills with AI

https://conversly.space
1•Kehvinbehvin•21m ago•0 comments

Surprise snowfall blankets the driest desert: Atacama turns white

https://www.ynetnews.com/environment/article/hjxnwy2egg
1•howard941•21m ago•0 comments

A Snapshot from the Walkman Era

https://ertu.dev/posts/a-snaphot-from-the-walkman-era/
3•ertucetin•22m ago•1 comments

Show HN: AI-SDK-Cpp – Unified C++ SDK for OpenAI, Anthropic, and More

https://github.com/iskakaushik/ai-sdk-cpp
2•cauchyk•35m ago•0 comments

Official Brave F-Droid repository now available

https://brave.com/blog/f-droid/
3•LorenDB•35m ago•0 comments

Theoretical Analysis of Positional Encodings in Transformer Models

https://arxiv.org/abs/2506.06398
2•PaulHoule•35m ago•0 comments

Rust in the Linux kernel: part 2

https://lwn.net/SubscriberLink/1025232/fbb2d90d084368e3/
2•chmaynard•35m ago•0 comments

First-Ever SpaceX Shares Now Available Through Blockchain

https://www.forbes.com/sites/digital-assets/2025/06/26/first-ever-spacex-shares-now-available-through-blockchain/
2•mywacaday•36m ago•0 comments

Leveraging (abusing?) external formatters in Zed

https://adamhl.dev/blog/zed-external-formatters/
1•genshii•36m ago•0 comments

Programmatic SEO as Early-Growth Infrastructure for B2B SaaS Startups

https://guptadeepak.com/the-early-stage-growth-trap-how-smart-startups-escape-the-marketing-catch-22/
1•guptadeepak•38m ago•1 comments

The Complete Rust Security Handbook

https://github.com/yevh/rust-security-handbook
2•Ganipote•38m ago•0 comments

Show HN: Learnings from an automated grant writing app developer

https://harryherzberg.substack.com/p/lessons-from-building-changefinder
1•spiderjew•42m ago•0 comments

Shadow of a Doubt

https://harpers.org/archive/2025/07/shadow-of-a-doubt-ocd-andrew-kay/
1•samclemens•42m ago•0 comments

Sniffnet v1.4: it's 2X faster than Wireshark at processing PCAP files

https://sniffnet.net/news/v1.4/
3•todsacerdoti•42m ago•0 comments

Rhythm Is All You Need – Goodbye Transformers Rhythm Is the Final Piece

https://medium.com/@rikudatian0/rhythm-is-all-you-need-the-age-of-transformers-is-over-d5fd755ed5fa
2•RyukuLogos•46m ago•0 comments