
AI thought follower @ramplabs
New York, NY
Joined January 2025
- Tweets810
- Following201
- Followers611
- Likes5.3K
Alex L retweeted
We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math.
It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training.
We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵
Blog: rikiyat.github.io/blog/weigh…
Readers added context they thought people might want to know
LLM was given weights explicitly as text in prompt and used code/forward passes (per authors' logs) to ablate Qwen looping direction, a known thinking-mode issue per model card. Primarily fixed repetition in that mode, not general 'natively' improving weights/math.
rikiyat.github.io/blog/weight-sp…
github.com/RikiyaT/lm_und…
huggingface.co/Qwen/Qwen3.5-0…
huggingface.co/RikiyaT/Qwen3.…
Introducing Ramp Accounting Bench.
We partnered with accounting professionals to create 137 tasks and grading rubrics grounded in everyday accounting workflows.
Even with three attempts, the best model achieved only 21% accuracy. Reliable agentic accounting remains an open challenge.
Alex L retweeted
Monitor and control your AI spend on every provider on router.com. Our early users save 40% on average.
Every week, the price-intelligence-latency frontier shifts, and we expect this trend to continue. Tradeoffs between latency, reasoning, cost, service tier, open source and closed source models are shifting constantly.
Router sends every request to the model that's actually best for the task and helps you control what tokens you buy. We benchmark it against real work: ~40% lower cost for the same outputs.
Today we're opening it to everyone.
Two lines of code or just change your base URL. No @tryramp account needed.
Free through 2026, first $26 on us.
Get an API key today at router.com
We just released an oss version of PorTAL, including training scripts and links to the paper’s models on Hugging Face: github.com/ramp-public/porta… Looking for alpha testers and feedback, give it a spin and let us know what you think!
Alex L retweeted
What causes #poverty? Nothing. It's the original state, the default and starting point. The real question is, What causes #prosperity?
so many intelligent people still cling to the idea that everything the ai labs do is downstream of IPO hype or whatever. some kinds of intelligentsia are uniquely unsuited for the moment because their lens is “strategic skepticism” where normal people can employ magical thinking
Alex L retweeted
Introducing Stack.
The AI operating system that lets accounting firms take on more clients without hiring. Learns your firm's process, runs the close, posts the journals. Fully auditable.
We’re living through the biggest shift in accounting since the spreadsheet.
Alex L retweeted
bro it isn’t generally intelligent bro its only read every book and paper ever written and just making connections between them bro. its only thinking for twenty hours bro it’s just brute force thinking bro. its only solving erdos problems bro it could never be an accountant bro
Alex L retweeted
.@RampLabs (the AI unit of @tryramp) has been *cooking* with agentic innovation
Here's @a_levitator discussing and demo'ing code self-maintaining software and the concept of AI software factories #DataDrivenNYC
______________
00:04 - Intro
01:11 - The shift from writing code to code maintenance
01:59 - Introducing Ramp Inspect, the background coding agent
03:05 - The first experiment: Nightly AI code automation
04:23 - The limits of stateless monitoring in large observability surfaces
05:47 - Using Datadog monitors to give the AI state and focus
07:23 - Real-world example: AI autonomously fixing an authentication bug
08:14 - How to control noise and implement an AI triage pattern
09:27 - The old vs. new paradigm for continuous code observability
10:21 - Key learnings on building autonomous AI software factories
Alex L retweeted
AI folks in NYC -- Data Driven NYC (#121) this Tuesday at 6pm.
Come meet fellow AI builders and our speakers:
* @RampLabs has been cooking lately with a lot of agentic innovation; Alex Levinson will demo
* @EstuaryDev provides unified data infra for AI - CEO @dyaffe
RSVP: luma.com/ddnyc121
We built a codebase that maintains itself.
An agent instruments every pull request, triages alerts, and pushes fixes autonomously. The system runs on a thousand AI-generated monitors, one for every 75 lines of code.
Alex L retweeted
Replying to @RampLabs
@ramplabs is where we experiment with llms at @tryramp. we run long-term projects like ramp.com/sheets, and one-offs like jay sobel’s claude rct + llm games.
we’re expanding the team and looking for full-stack engineers who love to tinker.
apply: jobs.ashbyhq.com/ramp?utm_so…
Alex L retweeted
yes things are changing fast, but also I see companies (even faang) way behind the frontier for no reason.
you are guaranteed to lose if you fall behind.
the no unforced-errors ai leader playbook:
For your team:
- use coding agents. give all engineers their pick of harnesses, models, background agents: Claude code, Cursor, Devin, with closed/open models. Hearing Meta engineers are forced to use Llama 4. Opus 4.5 is the baseline now.
- give your agents tools to ALL dev tooling: Linear, GitHub, Datadog, Sentry, any Internal tooling. If agents are being held back because of lack of context that’s your fault.
- invest in your codebase specific agent docs. stop saying “doesn’t do X well”. If that’s an issue, try better prompting, agents.md, linting, and code rules. Tell it how you want things. Every manual edit you make is an opportunity for agent.md improvement
- invest in robust background agent infra - get a full development stack working on VM/sandboxes. yes it’s hard to set up but it will be worth it, your engineers can run multiple in parallel. Code review will be the bottleneck soon.
- figure out security issues. stop being risk averse and do what is needed to unblock access to tools.
in your product:
- always use the latest generation models in your features (move things off of last gen models asap, unless robust evals indicate otherwise). Requires changes every 1-2 weeks - eg: GitHub copilot mobile still offers code review with gpt 4.1 and Sonnet 3.5 @jaredpalmer. You are leaving money on the table by being on Sonnet 4, or gpt 4o
- Use embedding semantic search instead of fuzzy search. Any general embedding model will do better than Levenshtein / fuzzy heuristics.
- leave no form unfilled. use structured outputs and whatever context you have on the user to do a best-effort pre-fill
- allow unstructured inputs on all product surfaces - must accept freeform text and documents. Forms are dead.
- custom finetuning is dead. Stop wasting time on it. Frontier is moving too fast to invest 8 weeks into finetuning. Costs are dropping too quickly for price to matter. Better prompting will take you very far and this will only become more true as instruction following improves
- build evals to make quick model-upgrade decisions. they don’t need to be perfect but at least need to allow you to compare models relative to each other. most decisions become clear on a Pareto cost vs benchmark perf plot
- encourage all engineers to build with ai: build primitives to call models from all code bases / models: structured output, semantic similarity endpoints, sandbox code execution. etc
What else am I missing?
I've never felt this much behind as a programmer. The profession is being dramatically refactored as the bits contributed by the programmer are increasingly sparse and between. I have a sense that I could be 10X more powerful if I just properly string together what has become available over the last ~year and a failure to claim the boost feels decidedly like skill issue. There's a new programmable layer of abstraction to master (in addition to the usual layers below) involving agents, subagents, their prompts, contexts, memory, modes, permissions, tools, plugins, skills, hooks, MCP, LSP, slash commands, workflows, IDE integrations, and a need to build an all-encompassing mental model for strengths and pitfalls of fundamentally stochastic, fallible, unintelligible and changing entities suddenly intermingled with what used to be good old fashioned engineering. Clearly some powerful alien tool was handed around except it comes with no manual and everyone has to figure out how to hold it and operate it, while the resulting magnitude 9 earthquake is rocking the profession. Roll up your sleeves to not fall behind.
Alex L retweeted
Technological innovation can be a form of participation in the divine act of creation. It carries an ethical and spiritual weight, for every design choice expresses a vision of humanity. The Church therefore calls all builders of #AI to cultivate moral discernment as a fundamental part of their work—to develop systems that reflect justice, solidarity, and a genuine reverence for life.
Alex L retweeted
one day someone at @tryramp just suggested this idea during Standup and now it's live in NYC
Alex L retweeted
Life is suffering, so choose something worth suffering for. -@GrahamCWeaver
Alex L retweeted
I've been surprised and impressed by how all of the leading agents are essentially while loops with llm + tool calls.
As a systems nerd, I've always wished that AI software (even pre-LLM) were modular. I love abstractions :)
But the reality is, the bitter lesson of AI systems is that the more you can push decision making & control flow to the model, the better your system will get as models improve. Therefore, the simplest architecture wins.





























