Expanse (YC P26) – Unlock Wasted GPU Capacity
Hey HN, we’re Ismaeel, Eren, Yafet and Nikodem. We built Expanse (https://expanse.sh/) to increase the effective capacity of your HPC/GPU clusters running schedulers/orchestrators like Kubernetes and SLURM. We read the source code, job submission script, and the hardware a workload is about to run on to predict what the job actually needs before the cluster sees it. We also flag failures we think are about to happen and surface line-level optimisations the researcher can apply themselves.The problem: Datacenters run at roughly 30% to 40% effective utilisation. Users request more resources than what they actually need, because of asymmetric risk: while over-requesting is bad because it’s expensive and wastes capacity that someone else could have used, under-requesting kills your job mid-run and you lose days of work. So everyone over-requests by two to three times.We measured one national-scale HPC cluster for a month and from 122k jobs, 59% of the compute was wasted. At on-demand cloud rates for the same hardware, that’s roughly $8.5M of compute wasted in one month on one cluster. The pattern is similar in large scale compute industries as well, such as quant funds, AI labs, and manufacturing.The four of us ran HPC and GPU training workloads at the largest quant funds and HPC facilities. Ismaeel did research at EPCC (Edinburgh’s Parallel Computing Centre, the UK’s national HPC site) under Adrian Jackson, where he built the first multimodal HPC resource predictor: a model that ingests job source code, submission scripts, hardware telemetry and cluster metadata in order to figure out how much compute will actually be needed. On a dataset of real workloads on EPCC’s own clusters it scored 34% better than any other baseline, and outperformed frontier general-purpose LLMs prompted on the same prediction task by roughly 8x. These results convinced us the problem was solvable with software.Expanse installs on every node and hooks into SLURM (or the
FL score
out of 100
Verdict
high confidence
Competition
No competitor data yet
Trend
No signal yet
Software that predicts actual GPU/compute resource needs before jobs run, recovering 30-40% of wasted datacenter capacity worth millions monthly per facility.
The pain
The gap
Build angle
Strengths
- Quantified problem: $8.5M waste monthly on one cluster, 59% of compute wasted, measured across 122k jobs
- Proven technical approach: 34% better than baselines, 8x better than prompted LLMs on real data
- Founder expertise: Team has run actual workloads at scale at EPCC and quant funds, not theoretical knowledge
- Low friction adoption: Installs on nodes, hooks into existing SLURM/Kubernetes, no workflow changes needed
- Multiple revenue angles: Per-job pricing, per-node licensing, percentage of savings, or enterprise contracts
- Massive TAM: Every HPC facility, cloud provider, quant fund, and AI lab has this problem
Risks
- Adoption requires cluster operators to trust predictions and change resource allocation policies; conservative IT orgs move slowly
- Prediction accuracy degrades on novel workloads or hardware the model hasn't seen; continuous retraining needed
- Kubernetes adoption in HPC is still growing; SLURM dominance means initial market is smaller than it appears
- Competitors could emerge: cloud providers (AWS, GCP) could build this internally; open-source alternatives could commoditize
- Pricing model unclear: if charged per job, users might resist; if percentage of savings, requires trust in measurement
- Integration complexity: each cluster has custom configurations, job types, and hardware; one-size-fits-all won't work
- Data privacy: analyzing job source code and cluster telemetry requires handling sensitive research/trading code
Fly Labs Method
Is the pain real, is there a gap, is it the right time, can one person build it.
Value Equation
Dream outcome and how likely it feels, against the time and effort it costs.
One-Person Business
Curiosity pull, identity fit, and a path from free value to paid for a solo creator.
Viral Frameworks
Hook strength, shareability, and how cheaply it can be tested.
Builder Lens
Evidence the problem exists, timing, defensibility, and a model that fits on a napkin.
Five lenses, one composite. How scoring works
The angle
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
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
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
Questions about this idea?
FlyBot reads the scoring and gives you a second opinion on “Expanse (YC P26) – Unlock Wasted GPU Capacity”.