Viewing profile — nicola_alessi

Shipped today. The benchmarks are real: 87.6% SWE-bench (from 80.8%), +13% on coding tasks, 3x more resolved production tasks on Rakuten-SWE-Bench. But there are a few changes that…

Fair point, appreciate the callout. I'll dial it back.

Interesting framing — hadn't thought about it from the inference routing angle but it maps well to what the data shows. On latency variance: yes, significantly. Cost standard devia…

I ran a controlled benchmark on AI coding agents (42 runs, FastAPI, Claude Sonnet 4.6) and found something that broke my mental model of LLM costs. The setup: I built an MCP server…

I built an MCP server (vexp) that pre-indexes a codebase into a dependency graph and serves only relevant code to AI coding agents. While benchmarking it, I found something I wasn'…

vexp works regardless of how you organize files. The graph is at the symbol level (functions, classes, types) not the file level, so whether you have one function per file or 50, i…

I got tired of watching Claude Code read entire files when it needed one function. Built an MCP server that pre-computes a dependency graph with tree-sitter and serves only the rel…

vexp is an MCP server that gives AI coding agents structural awareness of your codebase. Instead of letting the agent read entire files, it pre-computes a dependency graph with tre…

Gating completion on the observation write is smart — you're turning the model's task-completion drive against itself. Have you run into the problem where forced observations degra…

I've spent months building tooling for AI coding agents and hit something I can't fully explain. If you give an agent (Claude Code, Cursor, Codex) a tool to save observations — "sa…

I built vexp because AI coding agents have two expensive problems: they waste tokens reading irrelevant code, and they forget everything between sessions. The token problem: agents…

I built vexp to solve two problems I kept hitting with AI coding agents (Claude Code, Cursor, etc.): 1. Token waste: agents read entire files linearly to understand a codebase. On …

I've been building vexp for the past months to solve a problem that kept bugging me: AI coding agents waste most of their context window reading code they don't need. The problem W…

You are 100% correct, and this is the central limitation. An LLM like ChatGPT, trained on general web text, is a terrible movie recommendation engine for exactly the reasons you st…

This is a fantastic point, and you've hit on something fundamental that's been lost in the shift to on-demand: the joy of discovery through serendipity and low commitment. You're d…

Your point is excellent and cuts to the core of what we're trying to explore. You're right, ‘mood' can be a fuzzy, high-friction starting point. The hypothesis behind the prompt is…

For the last year, I've been obsessed with a problem: finding something to watch is a chore. The interfaces of Netflix, Prime, and others feel like slot machines designed for maxim…

I’ve been testing a platform called Lumigo.tv, and it made me rethink how recommendation systems could work if they started from human emotion instead of metadata. The core idea is…