Alternatives
Products that do what Semble – Code search for agents that uses 98% fewer tokens than grep does
Hey HN! We (Stephan and Thomas) recently open-sourced Semble. We kept running into the same problem while using Claude Code on large codebases: when the agent can't find something directly, it falls back to grep, reading full files or launching subagents. This uses a lot of tokens, and often still misses the relevant code. There are existing tools for this, but they were either too slow to index on demand, needed API keys, or had poor retrieval quality. So we built Semble. It combines static Model2Vec embeddings (using our latest static model: potion-code-16M) with BM25, fused via RRF and…
- 1SC
Hey HN! We (Stephan and Thomas) recently open-sourced Semble. We kept running into the same problem while using Claude Code on large codebases: when the agent can't find something directly, it falls back to grep, reading full files or launching subagents. This uses a lot of tokens, and often still misses the relevant code. There are existing tools for this, but they were either too slow to index on demand, needed API keys, or had poor retrieval quality. Semble is our solution for this. It combines static Model2Vec embeddings (using our latest static model: potion-code-16M) with BM25, fused…
May 2026 · github.com
- 2SF
Hey HN! We've just open-sourced Semble, a fast and accurate code search library built for agents. We're also releasing potion-code-16M, a small code-specialized static embedding model that powers it. Most embedding-based code search methods are either too slow to index on demand or need GPU infrastructure, while grep-style retrieval methods often cannot find the relevant content. Semble combines the speed and quality benefits of both, so agents waste less time and fewer tokens exploring. Main features: - Fast: indexes a full codebase in ~250 ms and answers queries in ~1.5 ms, all on CPU…
Apr 2026 · github.com
- 3
- 4
- 5
- 6
- 7
- 8CT
I've been building a tool that changes how LLM coding agents explore codebases, and I wanted to share it along with some early observations. Typically claude code globs directories, greps for patterns, and reads files with minimal guidance. It works in kind of the same way you'd learn to navigate a city by walking every street. You'll eventually build a mental map, but claude never does - at least not any that persists across different contexts. The Recursive Language Models paper from Zhang, Kraska, and Khattab at MIT CSAIL introduced a cleaner framing. Instead of cramming everything into…
Feb 2026 · github.com
- 9
- 10IN
Tl;dr: I trained a classifier to route to the least expensive model and reasoning depth to complete the request. Coupling that with additional automated token efficiency techniques has yielded 3x usage for the same spend. For anyone interested in trying it themselves: https://nerfguard.com Various teammates and I switched over to Codex from Claude Code recently. We still bounce between the tools, but Codex’s speed and steerability coupled with performance gains were hard to ignore. One of the downsides was that the per token pricing kicked in way sooner. This is happening across…
Jun 2026
- 11TI
I'm an "ideas person" who messes around with AI on a low budget. I got tired of watching my tokens vanish and context windows filling up while agents fumbled around trying to find the right thing. Agents don't flail like they used to with shell tools, but there are still weak/blind spots and back-and-forth episodes — especially when using tools in combination/sequence. So I built "tilth" today. Or rather, AI built it — every line is Opus 4.6. I spent a lot of my precious tokens getting it to "not shit" (at least several of the different vendors' AI overlords assure me it's not…
Feb 2026 · github.com
- 12
Hi y'all. Been working on something that should've been made a long time ago imo. It compiles codebases into O(1) hashmaps that the agent queries to discover the structure of your code/answer questions/write code. It also does complete static analysis checks on any writes the agent makes. Don't take my word for it though. Here are the benchmarks: https://benzi.fly.dev/benchmark. on 2/20 tests, Claude Code (mostly Sonnet on one task) regressed or timed out. Benzi didn't because of course, it has a map it can query and not get lost in the sauce. On the other 18 it…
29d ago · benzi.fly.dev
- 13AB
Hi HN, Zidan here. I’ve been experimenting with AI-assisted debugging and noticed a recurring gap: most tools optimize for agent-led exploration (ex: giving claude code a browser to click around and try to reproduce an issue). But in many cases, I've already found the bug myself. What I actually want is a way to hand the agent the exact context I just saw - without retyping steps, copying logs, or hoping it can reproduce the behavior. So we built FlowLens, an open-source MCP server + Chrome extension that captures browser context and lets coding agents inspect it as structured, queryable…
Nov 2025 · github.com
- 14
Cut Claude Code usage: −49% on code, −75% on reads.
Jul 2026 · greenpt.com
- 15BC
We are a small group of undergrads interested in building human in the loop coding agents. We dream of a world where building complex agent workflows feels as simple and creative as playing with legos. When we were building stuff we needed a tool that made it easy to try out different code embedding models so that we could see which ones worked best in different scenarios and understand their strengths and weaknesses. So to speed that process up we made PurpleSearch an 'instant' search engine for your local codebases. This tool lets you quickly deploy any open source embedding model on…
2025
- 16SR
Hello all, I'm a software developer. Over the last few months more and more of my work has turned into using coding agents instead of typing the whole code myself. Usually a few claude sessions at once, sometimes codex, one per feature or per revealed bug. I ran them in a split terminal for a few weeks, and quickly spotted two main problems. The first is that I couldn't easily tell which agent was stuck waiting on me and which was still working, so I'd cycle through sessions and checking on them. The second one: agents sharing a single branch step on each other. Two of them could be editing…
Jul 2026 · shikigami.dev
- 17MA
This weekend I built a multi-agent coding system which, quite unexpectedly, beat Claude Code on Stanford's Terminal Bench! The architecture is straightforward, consisting of an orchestrator agent that deploys explorer & coder subagents to complete complex terminal based tasks, utilising an intelligent context sharing mechanism along the way which makes it all work. The repo has a lot of technical details, and all the code and prompts for you to play around with if you'd like! I had a lot of fun making this, I hope you have fun reading the README, using it yourself, or even extending it! As…
2025 · github.com
- 18FC
Mar 2026 · ory.com
- 19SM
I built this because I got tired of watching Claude Code read through massive files just to find a few functions. Sourcerer lets AI agents search code semantically and grab exactly the code chunks they need instead of burning tokens on whole files. It uses tree-sitter to parse your codebase and creates a searchable index. So instead of "read auth.py (538 lines)", an agent can search for "user authentication logic" and get back just the relevant functions. Demo: https://asciinema.org/a/736638 GitHub: https://github.com/st3v3nmw/sourcerer-mcp
2025 · github.com
- 20PR
Hi HN, While building RAG agents, I noticed a lot of token budget was wasted on formatting overhead (HTML tags, JSON structure, whitespace). Existing solutions felt too heavy (often requiring torch/transformers), so I wrote this lightweight, zero-dependency library to solve it. It includes strategies for context packing, PII redaction, and tool output compression. Benchmarks show it can save ~15% of tokens with negligible latency overhead (<0.5ms). Happy to answer any questions!
Dec 2025 · github.com
- 21CA
Codebased combines Tree Sitter for code awareness (find functions, data structures, constants, etc. not just lines of code), full-text search using SQLite, and semantic search using OpenAI embeddings + FAISS. Despite being implemented in Python, supporting semantic search, making multiple API calls for embedding and re-ranking, it is faster than ripgrep for runng searches against the Linux kernel (takes ~1 second vs. ~2 seconds, obviously depends on system, temperature, time of day, tidal forces, etc.) Up next: - A Perplexity-like agent for interpreting results, making multiple follow-up…
2024 · codebased.sh
- 22AA
Hi HN, Even the smartest AI coding agents stall when the fix isn’t in their training data. AgruSeek runs an agentic search loop across ~30 M developer sources to dig up solutions normal web search misses. REAL‑WORLD USES • Found an undocumented `--runtime‑bypass` flag (buried in a 2017 gist) • Pulled actual Claude Code pricing from forum anecdotes - no “contact us” paywalls • Traced a race condition by cross‑linking five issue trackers across forks WHY POST NOW We’ve abused AgruSeek internally for three weeks; we’d love outside stress tests. Access is free (limited seats for Beta, no…
2025 · agruseek.com
- 23
Hey HN! We built https://keenable.ai, a different web search API for AI agents. Keenable searches our own 100B+ page index. We are focused on low cost and latency (p95 <250ms from us-east). We don’t believe in benchmaxxing, so we open-sourced our internal benchmarking suite, NEEDLE (available at https://keenableai.github.io/needle): a live benchmark that compares Keenable with other search APIs on fresh agent-like queries. I spent seven years at Amazon as a scientist working on web grounding for Alexa/AGI, and my co-founder Andrey previously led search at…
12d ago · keenable.ai
- 24IB
The main goal of this was to be able to not just run multiple Claude Code sessions at once, but actually manage them and keep track of what I was doing. Sometimes this is multiple attempts on the same task, sometimes I work several tasks at once. Really I was just sick of twiddling my thumbs waiting for the coding agent to finish, and I wanted it to be easy to work on/review/test another change while I waited.
2025 · github.com
Ranked by how close each launch is in meaning, then by votes. Refine with a description →