Twill.ai (YC S25) – Delegate to cloud agents, get back PRs
Hey HN, we're Willy and Dan, co-founders of Twill.ai (https://twill.ai/). Twill runs coding CLIs like Claude Code and Codex in isolated cloud sandboxes. You hand it work through Slack, GitHub, Linear, our web app or CLI, and it comes back with a PR, a review, a diagnosis, or a follow-up question. It loops you in when it needs your input, so you stay in control.Demo: https://www.youtube.com/watch?v=oyfTMXVECbsBefore Twill, building with Claude Code locally, we kept hitting three walls1. Parallelization: two tasks that both touch your Docker config or the same infra files are painful to run locally at once, and manual port rebinding and separate build contexts don't scale past a couple of tasks.2. Persistence: close your laptop and the agent stops. We wanted to kick off a batch of tasks before bed and wake up to PRs.3. Trust: giving an autonomous agent full access to your local filesystem and processes is a leap, and a sandbox per task felt safer to run unattended.All three pointed to the same answer: move the agents to the cloud, give each task its own isolated environment.So we built what we wanted. The first version was pure delegation: describe a task, get back a PR. Then multiplayer, so the whole team can talk to the same agent, each in their own thread. Then memory, so "use the existing logger in lib/log.ts, never console.log" becomes a standing instruction on every future task. Then automation: crons for recurring work, event triggers for things like broken CI.This space is crowded. AI labs ship their own coding products (Claude Code, Codex), local IDEs wrap models in your editor, and a wave of startups build custom cloud agents on bespoke harnesses. We take the following path: reuse the lab-native CLIs in cloud sandboxes. Labs will keep pouring RL into their own harnesses, so they only get better over time. That way, no vendor lock-in, and you can pick a different CLI per task or combine them.When you giv
FL score
out of 100
Verdict
high confidence
Competition
12
competitors found, emerging market, funded players
Trend
No signal yet
A cloud platform to delegate coding tasks to lab-native AI agents in sandboxes, aiming to solve parallelization, persistence, and trust issues for developers.
The pain
The gap
Build angle
Strengths
- Addresses highly specific and severe developer pain points (parallelization, persistence, trust).
- Large and rapidly growing market for AI coding assistants.
- Clear willingness to pay evidenced by competitor pricing.
- Specific technical angle ('reuse lab-native CLIs in cloud sandboxes') that could offer flexibility.
- Founder-market fit, as the creators built it for their own pain.
Risks
- Extremely crowded market with well-funded competitors (Devin, Sweep AI) offering similar 'cloud agent' functionality.
- High technical complexity for a solo builder to create a robust and scalable sandboxing/orchestration platform.
- Need to prove that developers strongly value Twill's specific 'reuse lab-native CLIs' approach over integrated solutions.
- Potential for AI labs to integrate similar cloud orchestration directly into their offerings, eroding Twill's niche.
Next steps
- Conduct in-depth interviews with developers currently using local AI coding agents or competing cloud agents to understand if Twill's specific 'reuse lab-native CLIs' and 'user control' differentiators are highly valued.
- Validate pricing model and whether the proposed value overcomes switching costs from existing solutions (both local setups and cloud competitors).
- Focus MVP on the most critical pain point (e.g., persistent execution of a single popular LLM CLI for specific tasks) to accelerate market feedback and minimize initial build complexity.
- Explore partnerships or integrations with specific AI labs to ensure compatibility and ease of use for their CLIs.
Fly Labs Method
Is the pain real, is there a gap, is it the right time, can one person build it.
- Problem clarity
- 25
- Solution gap
- 14
- Willingness to pay
- 19
- Buildability
- 10
Twill.ai addresses a real and specific pain for developers struggling with local AI coding agents, leveraging existing LLM CLIs in the cloud for parallelization, persistence, and trust. However, the market is highly competitive with strong, funded incumbents offering similar solutions.
Value Equation
Dream outcome and how likely it feels, against the time and effort it costs.
This idea has a clear value proposition in a fast-growing market, but strong competition and significant build complexity for a solo builder pose challenges for profitability and defensibility.
One-Person Business
Curiosity pull, identity fit, and a path from free value to paid for a solo creator.
This project has high problem clarity and good monetization potential for developers, but it's a complex build with an 'anti-niche' focus that might be challenging for a solo builder to execute and differentiate.
Viral Frameworks
Hook strength, shareability, and how cheaply it can be tested.
Twill.ai has a clear target audience and value proposition for a viable micro-SaaS, with a proven business model, but faces high risk from market saturation and the need to validate its specific differentiation.
Builder Lens
Evidence the problem exists, timing, defensibility, and a model that fits on a napkin.
This idea addresses a real and growing problem for developers, with strong founder insights, but needs to prove its unique value proposition against many funded competitors.
Why this verdict
Five lenses, one composite. How scoring works
The angle
This weekend
- Conduct in-depth interviews with developers currently using local AI coding agents or competing cloud agents to understand if Twill's specific 'reuse lab-native CLIs' and 'user control' differentiators are highly valued.
- Validate pricing model and whether the proposed value overcomes switching costs from existing solutions (both local setups and cloud competitors).
- Focus MVP on the most critical pain point (e.g., persistent execution of a single popular LLM CLI for specific tasks) to accelerate market feedback and minimize initial build complexity.
- Explore partnerships or integrations with specific AI labs to ensure compatibility and ease of use for their CLIs.
Who is already there, emerging market
An AI pair programmer that provides autocomplete-style suggestions from an AI model trained on billions of lines of code.
Pricing: Individual: $10/month, Business: $19/user/month, Enterprise: $39/user/month.
An AI-first code editor, forked from VS Code, that integrates AI features directly into its core, including multi-file agent capabilities.
Pricing: Credit-based pricing; Enterprise teams needing centralized billing oversight pay $40/user/month.
Anthropic's terminal-based AI coding assistant, designed for deep context understanding and capable of handling complex software tasks.
Pricing: Starting at $20/month.
The world's first fully autonomous AI software engineer that can handle complex engineering tasks end-to-end in its own sandboxed environment.
Pricing: Core plan: starting at $20/month ($2.25 per Agent Compute Unit); Team plan: $500/month; Enterprise plan: Custom pricing.
An AI coding partner that automates entire workflows from issue to PR, understanding codebases, planning changes, writing code, and creating pull requests.
Pricing: Free Trial; Basic: $10/month; Pro: $20/month; Ultra: $60/month. Team plans include $40 in Sweep API credits per seat.
- Mutable.aiVisit
An AI-powered software development platform that accelerates coding workflows through contextual code generation, refactoring, and documentation.
Pricing: Free plan; Basic: $10/month; Pro: $30/month; Enterprise: Custom pricing.
- CodeRabbitVisit
Provides comprehensive AI-powered pull request reviews and advanced insights.
Pricing: Free tier for open-source projects; Pro: $24/month (billed annually) or $30/month (monthly) per developer.
- AiderVisit
An open-source, terminal-based pair programming tool that works with multiple LLM backends for flexible and controlled code edits.
Pricing: Free; open source (Apache-2.0). Pay only for the model/API if hosted.
Google's AI developer assistant offering conversational workflows and strong reasoning on large codebases, integrated with Google Cloud services.
Pricing: Free for individuals (1,000 requests per day at zero cost with a personal Google account).
OpenAI's terminal-native coding agent, rewritten in Rust for performance, running each task in a network-disabled cloud container for isolated execution.
Pricing: No direct pricing for Codex specifically, but OpenAI models generally have usage-based pricing; alternatives often cite its Pro subscription as expensive ($200/month).
- CosineVisit
An AI software engineer that operates across the CLI, cloud, and desktop, designed to complete tasks end-to-end without human supervision and open PRs.
Pricing: Pricing not publicly disclosed, but users praise its value for development teams.
- Kilo CodeVisit
An open-source, model-agnostic AI code assistant that provides fast, low-friction coding assistance in VS Code and JetBrains IDEs.
Pricing: Open-source with transparent pricing, pay-per-credit model for AI providers.
What they charge
Recent news
- Launch HN: Twill.ai (YC S25) – Delegate to cloud agents, get back PRs
Hacker News, April 10 2026
- The best AI coding agents in 2026
Product Hunt, April 10 2026
- 14 Best AI Code Review Tools in 2026 — Pricing & Features Compared
Git AutoReview, April 09 2026
- AI Coding Assistant Statistics 2026: 50+ Key Data Points
Uvik Software, April 08 2026
- AI Development Cost 2026: Complete Pricing Guide
Boundev, April 07 2026
Market signals
The AI code assistant market is experiencing significant growth, projected to reach $6.55 billion by 2035 with a CAGR of 5.33% from 2025-2035. Other estimates place the broader AI code tools market (including code generation, review, and testing) at $7–$10 billion in 2025–2026, with a projection to hit $47.3 billion by 2034, growing at a CAGR of 24% from 2025 to 2034. This indicates a rapidly expanding market driven by increasing developer productivity and automated code creation.
What frustrates people
- High subscription costs for some AI coding tools, such as OpenAI Codex's $200 monthly Pro subscription.
- AI-generated code that is overly complex, difficult to maintain, or challenging to integrate with existing projects.
- Generative AI limitations resulting in less precise solutions or the AI looping on complex refactors without making progress.
- Noise and irrelevant feedback from AI code review tools requiring significant configuration time to dial down.
- Lack of deeper codebase context and customizability for enterprise workflows in some tools.
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 “Twill.ai (YC S25) – Delegate to cloud agents, get back PRs”.