What is one simple thing LLMs are insanely bad at?: SKIP, 42/100 | Ideas Lab

Build a specialized LLM that solves one specific task that general models consistently fail at.

The pain

Users repeatedly hit walls with ChatGPT and Claude on particular tasks, counting tokens accurately, maintaining state across conversations, following complex multi-step instructions, or handling specific domain logic. They waste time working around these failures or switching tools.

The gap

General-purpose LLMs optimize for breadth, not depth. They're trained to be okay at everything, which means they're mediocre at many specific things. No one has built a focused model that dominates one narrow task that matters to paying customers.

Build angle

Start by identifying which specific failure mode appears most frequently across users and has the highest switching cost. Examples: accurate JSON parsing, reliable math operations, consistent instruction following, or domain-specific reasoning. Build a model fine-tuned specifically for that task, then charge based on accuracy improvement or API calls.

Strengths

Clear problem space, LLM limitations are observable and measurable
Low barrier to entry, can fine-tune existing models rather than train from scratch
Potential for high margins, specialized tools command premium pricing
Easy to validate, can survey users or analyze error logs to find the most painful failure mode

Risks

Problem identification is vague, the idea is a question, not a solution. You haven't identified which specific task to target yet
Commoditization risk, as base models improve, your specialized advantage erodes quickly
Scaling challenge, a single-task model has limited TAM unless that task is used by millions
Competitive pressure, larger AI labs can add the same capability to their models faster than you can build a business
Unclear monetization, unclear if users will pay for a specialized model vs. using free general models with workarounds

Fly Labs Method

Is the pain real, is there a gap, is it the right time, can one person build it.

42

Value Equation

Dream outcome and how likely it feels, against the time and effort it costs.

35

One-Person Business

Curiosity pull, identity fit, and a path from free value to paid for a solo creator.

55

Viral Frameworks

Hook strength, shareability, and how cheaply it can be tested.

38

Builder Lens

Evidence the problem exists, timing, defensibility, and a model that fits on a napkin.

48

Five lenses, one composite. How scoring works

The angle

Start by identifying which specific failure mode appears most frequently across users and has the highest switching cost. Examples: accurate JSON parsing, reliable math operations, consistent instruction following, or domain-specific reasoning. Build a model fine-tuned specifically for that task, then charge based on accuracy improvement or API calls.

No market research recorded for this idea yet.

Semble – Code search for agents that uses 98% fewer tokens than grep

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 via RRF and reranked with code-aware signals. Everything runs on CPU since there's no transformers involved. On our benchmark of ~1250 query/document pairs across 63 repos and 19 languages, it uses 98% fewer tokens than grep+read and reaches 99% of the retrieval quality of a 137M-parameter code-trained transformer, while being ~200x faster.Main features:- Token-efficient: 98% fewer tokens than grep+read- Fast: ~250ms to index a typical repo on our benchmark, ~1.5ms per query on CPU (very large repos may take longer)- Accurate: 0.854 NDCG@10, 99% of the best transformer setup we tested- MCP server: drop-in for Claude Code, Cursor, Codex, OpenCode- Zero config: no API keys, no GPU, no external servicesInstall in Claude Code with: claude mcp add semble -s user -- uvx --from "semble[mcp]" sembleOr check our README for other installation instructions, benchmarks, and methodology:Semble: https://github.com/MinishLab/sembleBenchmarks: https://github.com/MinishLab/semble/tree/main/benchmarksModel: https://huggingface.co/minishlab/potion-code-16MLet us know if you have any feedback or questions!

AI

BUILD

85FL score

Postgres extension for BM25 relevance-ranked full-text search

Last summer we faced a conundrum at my company, Tiger Data, a Postgres cloud vendor whose main business is in timeseries data. We were trying to grow our business towards emerging AI-centric workloads and wanted to provide a state-of-the-art hybrid search stack in Postgres. We'd already built pgvectorscale in house with the goal of scaling semantic search beyond pgvector's main memory limitations. We just needed a scalable ranked keyword search solution too.The problem: core Postgres doesn't provide this; the leading Postgres BM25 extension, ParadeDB, is guarded behind AGPL; developing our own extension appeared daunting. We'd need a small team of sharp engineers and 6-12 months, I figured. And we'd probably still fall short of the performance of a mature system like Parade/Tantivy.Or would we? I'd be experimenting long enough with AI-boosted development at that point to realize that with the latest tools (Claude Code + Opus) and an experienced hand (I've been working in database systems internals for 25 years now), the old time estimates pretty much go out the window.I told our CTO I thought I could solo the project in one quarter. This raised some eyebrows.It did take a little more time than that (two quarters), and we got some real help from the community (amazing!) after open-sourcing the pre-release. But I'm thrilled/exhausted today to share that pg_textsearch v1.0 is freely available via open source (Postgres license), on Tiger Data cloud, and hopefully soon, a hyperscalar near you:https://github.com/timescale/pg_textsearchIn the blog post accompanying the release, I overview the architecture and present benchmark results using MS-MARCO. To my surprise, we were not only able to meet Parade/Tantivy's query performance, but exceed it substantially, measuring a 4.7x advantage on query throughput at scale:https://www.tigerdata.com/blog/pg-textsearch-bm25-fu

AI

BUILD

85FL score

Consus (YC F24) – LLM gateway for companies handling sensitive data

Hey Hacker News! I'm Eric, the founder of Consus. We route all AI requests to ITAR restricted and controlled unclassified information (CUI) compliant infrastructure. Consus itself satisfies ITAR restrictions, and we recently completed our FedRAMP equivalency audit against the 410 FedRAMP High controls. FedRAMP High is the U.S. government’s strictest security baseline for cloud services handling sensitive unclassified data. A breach of this data could cause severe or catastrophic harm. Achieving FedRAMP equivalency, or authorization, is required before processing CUI information.Demo: https://youtu.be/6MYxJe6P4oYAerospace and defense is experiencing rapid growth. Unfortunately, most companies are struggling to use AI because of necessary data protections & compliance risk. These companies can move even faster if they could use the AI tools that most other industries take for granted.I spent most of my career prior to Consus in aerospace & defense. I launched national security payloads at SpaceX, and was an early employee at Castelion. I’ve watched first hand how compliance red-tape makes it difficult for companies to adopt new technologies and trust vendors that touch their sensitive data. For example, I worked with a satellite manufacturer that doesn’t allow any AI tools, even for non-sensitive work.Even as AI has become extremely useful over the past 9-12 months, I was surprised by how many aerospace & defense companies refused to touch it. They could see how it could be helpful, but they were scared about model providers training on their data, violating an ITAR rule, or risk losing new business because of a failed cyber assessment. These are all rational fears.There were a lot of ways to solve some/all of these problems, but we focused on building a compliant AI inference layer that can "plug" into any AI tool or workflow. Just as OpenRouter integrates with most tools, Consus can connect to tools like ChatGPT or Clau

AI

BUILD

85FL score

All AI ideas