Context.dev
Web data API for AI agents to scrape, search, crawl, and extract structured data from the live web.
Website: https://www.context.dev/
Cover Block
Public sources
| Field | Value |
|---|---|
| Name | Context.dev |
| Tagline | Web data API for AI agents to scrape, search, crawl, and extract structured data from the live web [Context.dev, October 2026] |
| Headquarters | San Francisco, United States [Crunchbase] |
| Founded | 2025 [Crunchbase] |
| Stage | Seed [Crunchbase] |
| Business Model | API / Developer Platform |
| Industry | Other |
| Technology | AI / Machine Learning |
| Geography | Global / Remote-First |
| Growth Profile | Venture Scale |
| Founding Team | Solo Founder [Y Combinator, October 2026] |
| Funding Label | Seed |
| Total Disclosed | ~$9,000,000 [Context.dev, October 2026] |
Links
Public sources
- Website: https://www.context.dev/
- Pricing: https://www.context.dev/pricing
- Docs: https://docs.context.dev/introduction
- Y Combinator: https://www.ycombinator.com/launches/SPy-context-dev-live-web-data-api-for-ai-agents
Executive Summary
PUBLIC Context.dev sells a web data API that lets AI agents and software scrape, search, crawl, and extract structured data from the live web, and it merits investor attention now because the company paired that developer-facing wedge with a disclosed $9 million seed round in October 2026 and reported more than 800 customers at that point [Context.dev, October 2026] [Runtime Wire, October 2026] [Y Combinator, October 2026]. The company was founded in 2025 by Yahia Bakour and emerged from Y Combinator's Summer 2026 batch; public materials indicate it started as Brand.dev before rebranding to Context.dev in 2026 as the product expanded from brand data into broader web extraction and live context for agents [Y Combinator, October 2026] [Contentedge, March 2026] [Context.dev, October 2026].
The product proposition is straightforward: one REST API for scraping URLs into LLM-ready Markdown, searching the live web, crawling full sites, extracting structured data, and monitoring changes, with adjacent surfaces for company and brand data that can reduce the need for separate point tools inside agent and RAG workflows [Context.dev, October 2026] [Context.dev]. Public traction claims remain mostly company-reported, but they are directionally notable, including customer references such as Mintlify, Similarweb, SiteGPT, Passionfroot, and Rho, alongside a pricing model that starts with self-serve plans from $19 per month and scales to enterprise sales [Y Combinator, October 2026] [Silicon Valley Post, October 2026] [Context.dev].
On team quality, the public record is still founder-centric, which is typical at this stage. Bakour's LinkedIn shows prior engineering experience at Amazon from June 2020 to August 2022, and other public profiles tie him to Stock Alarm, a startup later sold through Acquire.com, which gives some evidence of product shipping and founder-market persistence even if the current bench appears small at roughly six to seven employees in 2026 sources [Yahia Bakour, LinkedIn, retrieved 2026] [Acquire.com Blog, February 2026] [Y Combinator, October 2026] [RocketReach].
Capital formation is unusually strong for a young infrastructure company. Context.dev's October 2026 seed was led by General Catalyst with participation from Orange Collective, Pioneer Fund, Robinhood Ventures, Eniac Ventures, SV Angel, Y Combinator, and angels including Paul Graham, Guillermo Rauch, Theo Browne, and Rick Rubin, which suggests the company has drawn attention from both institutional and operator-investor circles before a broader market proof point is fully public [Context.dev, October 2026] [Silicon Valley Post, October 2026].
The next 12 to 18 months turn less on narrative and more on proof. Investors should watch whether customer count converts into durable spend expansion, whether performance claims against protected sites continue to hold up in third-party testing, and whether the company can widen differentiation beyond a useful developer API into a harder-to-replicate data or workflow advantage in the agent stack [String, 2026] [favtutor.com, 2026] [Context.dev, October 2026].
Lightly corroborated -- This section relies on a mix of company announcements, Y Combinator materials, and a limited set of independent public sources; several traction and product claims remain company-reported.
Taxonomy Snapshot
| Axis | Value |
|---|---|
| Stage | Seed |
| Business Model | API / Developer Platform |
| Industry / Vertical | Other |
| Technology Type | AI / Machine Learning |
| Geography | Global / Remote-First |
| Growth Profile | Venture Scale |
| Founding Team | Solo Founder |
| Funding | Seed, total disclosed about $9,000,000 |
How the Company Got Here
PUBLIC
Context.dev is a young developer infrastructure company aimed at a current pain point: turning the live web into structured inputs that AI agents can reliably use. The company is headquartered in San Francisco and was founded in 2025, according to its Crunchbase profile [Crunchbase]. On its website, Context.dev describes itself as a "web data API for AI agents" that can scrape URLs, search the live web, crawl sites, and extract structured data through a single REST API [Context.dev, October 2026].
The public timeline is short but reasonably clear. In May 2026, the company announced that it was backed by Y Combinator, which places it in the accelerator's 2026 cohort and marks its first disclosed institutional program affiliation [Context.dev, May 2026]. By October 2026, the company had publicly announced a $9 million seed round led by General Catalyst and described a broader product scope under the Context.dev name [Context.dev, October 2026].
One milestone matters because it clarifies how management sees the product's evolution. The company states that it rebranded from Brand.dev to Context.dev as it expanded from brand and company data into a wider web-data API for agents and software [Context.dev, October 2026]. Crunchbase also lists founder Yahia Bakour as CEO, consistent with the company's own materials identifying him as the founder behind the product [Crunchbase].
Lightly corroborated -- Confirmed in part by Crunchbase and Context.dev materials, but several core company details in this section rely on company disclosures rather than independent public records.
Product and Technology
MIXED
The product is easiest to understand as a single API layer for live web retrieval and transformation, aimed at developers building AI agents and web-aware software. On its homepage and docs, Context.dev says users can scrape individual URLs into Markdown, search the live web, crawl full sites, map domains to discover URLs before scraping, run large batches, monitor pages for changes, and extract structured outputs through one REST API [Context.dev, October 2026] [Context.dev]. The same public materials also describe outputs that include Markdown, rendered HTML, screenshots, company profile data, brand assets, and customer-defined JSON, which suggests the company is trying to collapse several adjacent workflows, scraping, crawling, enrichment, and monitoring, into one developer surface rather than selling point products separately [Context.dev, October 2026] [Context.dev].
The public evidence also shows a product that has widened from its earlier Brand.dev identity into a broader web-context layer. Context.dev said in its October 2026 financing post that it expanded from brand data into infrastructure for agents to interact with the live web, and third-party coverage from Contentedge in March 2026 also reported the Brand.dev to Context.dev rebrand [Context.dev, October 2026] [Contentedge, March 2026]. Pricing is already packaged for self-serve adoption, with Developer at $19 per month, Pro at $99, Growth at $299, Scale at $499, plus enterprise plans, and the docs show integrations spanning SDKs, CLI, MCP, and external tools such as Zapier, Make, Google Sheets, Excel, Eve, and Composio [Context.dev, Unknown] [Context.dev, October 2026] [Composio]. That combination matters because it points to a developer-first distribution motion: low-friction entry on price, then expansion into broader use cases if the API becomes embedded in agent or data pipelines.
Performance claims should be treated with some care because the strongest numbers remain company-issued, but there is at least partial outside testing. Context.dev says it can scrape any URL with a 99.96% success rate, while third-party tests cited in the research found an 87.09% success rate in one web scraping API benchmark and pass rates of 90.7% on Cloudflare-protected targets, 83.0% on Akamai-protected targets, and 33.3% on Kasada-protected targets in another comparison [Context.dev, October 2026] [favtutor.com] [String]. Taken together, the product appears technically credible for mainstream scraping and retrieval workloads, but anti-bot resilience is uneven across protection layers and the section's most ambitious capability framing still rests largely on company materials rather than independent demos or technical audits.
Company-stated, unverified -- Material product capabilities and pricing are primarily confirmed by company website and docs, with limited third-party corroboration from Contentedge, Composio, favtutor.com, and String.
Where the Demand Sits
PUBLIC
The market matters now because AI agents have moved from static knowledge retrieval toward live web interaction, and that shifts infrastructure demand from model access alone to fresh, structured, machine-readable web data [Y Combinator, October 2026] [Context.dev, October 2026].
Public source support for a formal TAM, SAM, or SOM is thin in the materials reviewed, so the market has to be framed through adjacent, observable demand rather than top-down sizing. Context.dev positions itself as a single API for scraping, crawling, search, extraction, and monitoring across the live web [Context.dev, October 2026]. That positioning sits at the intersection of several existing software budgets: web scraping infrastructure, data enrichment, retrieval systems for RAG and agents, monitoring tools, and developer APIs for workflow automation [Context.dev, October 2026] [Context.dev].
The strongest demand signal in the cited record is not a third-party market report, but visible customer behavior across agent and research workflows. Y Combinator's launch profile describes use cases that range from turning repositories into documentation, to monitoring hundreds of healthcare and financial sources, to discovering and structuring API documentation across thousands of integrations [Y Combinator, October 2026]. The company's own October 2026 funding announcement says it had more than 800 customers, up from more than 400 in Y Combinator launch materials, which suggests buyer interest is broadening across developer and product teams building web-aware software [Context.dev, October 2026] [Y Combinator, October 2026].
A practical way to size the opportunity is to map the adjacent markets this product substitutes for. The nearest substitute is point-solution scraping infrastructure, where users historically combined scraping APIs, browser automation, search tools, enrichment vendors, change-detection systems, and internal ETL to keep external data usable for applications [Context.dev, October 2026]. A second adjacent market is enterprise data enrichment, especially for company, brand, and website intelligence, which fits Context.dev's origins as Brand.dev and its continued support for company profiles, brand assets, and taxonomy lookup tools [Contentedge, March 2026] [Context.dev, October 2026]. A third adjacent market is AI retrieval infrastructure, where freshness and source attribution matter more as agent products move beyond static vector stores into real-time web actions [Context.dev, October 2026] [Y Combinator, October 2026].
The tailwinds are clear enough even without a clean market-sizing deck. One is agent adoption itself: the homepage and launch materials are built around use cases such as research agents, monitoring, live search, grounding, and automatic personalization, which are all categories that break if external data is stale or poorly structured [Context.dev, October 2026] [Y Combinator, October 2026]. Another is workflow consolidation. If a developer can replace several brittle services with one API, procurement and implementation friction can fall, particularly for smaller teams shipping quickly inside the current agent build cycle [Context.dev, October 2026]. The company's pricing ladder, starting at $19 per month and moving to enterprise plans, also suggests the addressable demand spans individual builders through larger software teams, even if public evidence does not yet establish enterprise contract mix [Context.dev].
Macro and regulatory forces cut both ways. On the favorable side, the broader software market is rewarding tools that make AI systems more current, auditable, and operationally useful, especially where outputs need cited sources or ongoing monitoring [Context.dev, October 2026]. On the constraining side, any company built on live web extraction faces website defenses, platform policy shifts, and an uneven technical environment shaped by providers such as Cloudflare, Akamai, and Kasada; third-party testing cited in the research shows materially different pass rates across those environments, which underscores that this market is partly a distribution and reliability contest, not only a developer-experience one [String, 2026].
| Market lens | What the cited evidence supports |
|---|---|
| Core market | Web data infrastructure for AI agents and web-aware software [Context.dev, October 2026] [Y Combinator, October 2026] |
| Adjacent market 1 | Scraping and crawling APIs used to turn websites into machine-readable data [Context.dev, October 2026] |
| Adjacent market 2 | Company and brand data enrichment, reflecting the Brand.dev origin [Contentedge, March 2026] [Context.dev, October 2026] |
| Adjacent market 3 | Retrieval and monitoring infrastructure for RAG, research, and agent systems [Context.dev, October 2026] [Y Combinator, October 2026] |
| Demand evidence | More than 400 customers in YC launch materials, more than 800 customers in the October 2026 funding post [Y Combinator, October 2026] [Context.dev, October 2026] |
The table points to a market that is easier to validate through workflow substitution than through a neat top-down number. The important question is less whether web data for agents is real, public evidence says it is, and more whether one vendor can hold performance and compliance together as the category scales.
Lightly corroborated -- Relies primarily on company and Y Combinator materials for market framing, with limited independent third-party market sizing and one external performance source.
Competitive Landscape
Positioning
MIXED Context.dev is positioning itself less as a point scraping tool and more as a single API layer for live web retrieval, structured extraction, brand data, and monitoring workflows aimed at developers building AI agents and web-aware software [Context.dev, October 2026] [Y Combinator, October 2026].
| Company | Positioning | Stage / Funding | Notable Differentiator | Source |
|---|---|---|---|---|
| Context.dev | Web data API for AI agents that scrapes, searches, crawls, and extracts structured data from the live web | Seed, $9M disclosed in October 2026 | Broad workflow surface in one REST API, including scraping, live web search, monitors, answers, and brand/company data | [Context.dev, October 2026] [Y Combinator, October 2026] |
| Firecrawl | Named alternative for website crawling and scraping in agent workflows | Not established in the provided public source set | SiteGPT reportedly migrated its crawling from Firecrawl to Context.dev in under a day, which suggests direct overlap in crawl-to-Markdown use cases | [Context.dev, May 2026] [Context.dev, October 2026] |
The segment map is narrower than the company's broad product language first suggests. In direct competition, Context.dev appears to overlap most clearly with developer-first crawling and scraping APIs such as Firecrawl, especially where the job is turning websites into Markdown or structured outputs for retrieval, agents, or knowledge bases [Context.dev, May 2026] [Context.dev, October 2026]. Around that core are adjacent substitutes rather than fully verified direct peers in this source set: internal scraping stacks, pieced-together combinations of crawlers plus enrichment vendors, and search-based workflows that do not maintain a persistent web data layer. The company's own pitch is that those functions can be collapsed into one API, which is credible as a product ambition, though still mostly supported by company and accelerator materials rather than independent category reporting [Context.dev, October 2026] [Y Combinator, October 2026].
The edge that is most visible today is product breadth paired with early developer distribution. Public materials show one API surface spanning scrape, search, crawl, monitoring, file ingestion, and brand data, plus published pricing tiers and integrations that make the product legible to self-serve developers rather than only enterprise buyers [Context.dev, October 2026] [Context.dev]. Y Combinator backing and a General Catalyst-led seed round should help with distribution and hiring, and the customer count progression from more than 400 in the YC launch materials to more than 800 in the October fundraise announcement suggests that the product is finding an audience quickly [Y Combinator, October 2026] [Context.dev, October 2026]. Still, that edge looks perishable unless it hardens into either clearly superior reliability, a proprietary corpus, or workflow lock-in, because API aggregation alone is usually easier for well-funded rivals to copy than a data asset or exclusive channel.
The main exposure is that a focused competitor can still win a narrower category on depth. Firecrawl is the clearest named example in the public record, and the fact pattern cuts both ways: SiteGPT's reported migration to Context.dev is a positive signal for switching ease, but it also implies that buyers in this segment may treat providers as relatively substitutable if the interface and output quality are close enough [Context.dev, May 2026]. More broadly, several of Context.dev's most ambitious claims, including keeping 20 million company records fresh or powering 3,000-plus integrations, remain company-linked or secondary-source claims in the available record, not independently corroborated operating facts [Context.dev, October 2026]. That matters competitively because if the buyer's core requirement is verified coverage, compliance, or consistently high extraction performance across protected sites, the burden of proof rises quickly.
The most plausible 18-month scenario is a split market. Context.dev is the likely winner if developers keep preferring a single vendor that can move from crawl to extraction to monitoring without stitching together multiple tools, and if the company converts early self-serve usage into standard placement inside agent frameworks and product teams [Context.dev, October 2026] [Context.dev]. Firecrawl is the likely winner if the category settles around one primary job, website crawling for LLM pipelines, and buyers reward specialist reliability over breadth. In that case, the loser is whichever platform fails to turn usage growth into durable integration depth, because the switching evidence in the public record suggests customers can move when setup friction is low [Context.dev, May 2026] [Y Combinator, October 2026].
Lightly corroborated -- This section relies on two independent public sources for Context.dev's positioning and funding, but the direct competitor set is thin in the provided record and several comparative claims remain company- or accelerator-sourced [Context.dev, October 2026] [Y Combinator, October 2026] [Context.dev, May 2026].
Opportunity
PUBLIC
If Context.dev executes, the prize is not a niche scraping tool but a default data access layer for AI agents that need the live web, structured outputs, and deployment-ready infrastructure in one API [Context.dev, October 2026] [Y Combinator, October 2026].
The headline opportunity is straightforward: own the application-facing layer between AI software and the open web. That sounds expansive, but the public evidence makes it at least reachable. Context.dev is positioning beyond single-purpose scraping into search, site crawling, monitoring, brand data, structured extraction, and file ingestion through one REST API, which is a broader product surface than a point solution built around one workflow [Context.dev, October 2026] [Context.dev]. The early customer count also matters here. Y Combinator described the company at more than 400 customers in its launch materials, and the company said it had more than 800 customers by its October 2026 seed announcement, alongside named deployments such as Mintlify, Similarweb, SiteGPT, Passionfroot, Rho, and Zero [Y Combinator, October 2026] [Context.dev, October 2026] [Silicon Valley Post, October 2026]. For an API business founded in 2025, that is enough to suggest real pull from developers building agentic products, not just an interesting demo [Y Combinator, October 2026] [Crunchbase].
The more ambitious version of that outcome is that Context.dev becomes the standard abstraction layer developers reach for when an agent needs current information from websites, PDFs, company pages, or full domains. The case for that rests on product breadth and evidence of switching. The homepage and docs show one platform spanning scraping, web search, monitors, batch crawling, domain mapping, answers with sources, and extraction into Markdown or customer-defined JSON [Context.dev, October 2026] [Context.dev]. The company also cited SiteGPT moving its crawling off Firecrawl in under a day, which matters less as a single logo than as proof that migration friction may be low if reliability and breadth are good enough [Context.dev, May 2026] [Context.dev, October 2026].
| Scenario | What happens | Catalyst | Why it's plausible |
|---|---|---|---|
| Agent infrastructure default | Context.dev becomes the standard API that agent builders plug in for live web access, structured extraction, and monitoring across many use cases | A continued rise in agent products that need current web context rather than static training data | The platform already markets scrape, search, crawl, monitoring, answers, and brand data in one API, and reported more than 800 customers by October 2026 [Context.dev, October 2026] [Runtime Wire, October 2026] |
| Embedded inside AI toolchains | Context.dev wins distribution by sitting inside orchestration layers, no-code automation, and developer workflows | Integration-led adoption through SDKs and connectors | The company lists integrations for AI tools, agent frameworks, Zapier, Make, Google Sheets, Excel, and five official SDKs, which supports a credible distribution path beyond direct sales [Context.dev] |
| Data utility layer for enterprise apps | Context.dev expands from developer tooling into repeatable data services for research, enrichment, onboarding, and personalization products | Named customers prove category fit, then larger software vendors standardize on one vendor | Public materials cite use cases across Similarweb, Mintlify, SiteGPT, Zero, Soria, Murph, Dench, Passionfroot, and Rho, spanning research, documentation, CRM, monitoring, and data enrichment [Y Combinator, October 2026] [Context.dev, October 2026] |
What compounding looks like here is a product-and-distribution flywheel rather than a classical consumer network effect. Each new use case broadens the surface area of the API, which can raise retention if customers start with scraping and add search, monitors, answers, or brand data over time [Context.dev, October 2026] [Context.dev]. Each integration into developer workflows also lowers adoption friction for the next cohort of builders [Context.dev]. There is an early signal of this expansion dynamic in the rebrand from Brand.dev to Context.dev, which followed customer demand for broader website extraction and live web context, and in the pricing changes that raised plan quotas without a price increase, suggesting the company is trying to widen usage before fully optimizing monetization [Contentedge, March 2026] [Context.dev, October 2026] [Context.dev]. If reliability remains strong, third-party testing adds some support: String reported 90.7 percent success across Cloudflare-protected targets and 83.0 percent against Akamai-protected targets, while Favtutor reported an 87.09 percent success rate in a web scraping API test [String, 2026] [favtutor.com, 2026].
The size of the win is best framed through infrastructure value rather than near-term revenue. A credible public comparable is Similarweb, which operates in web intelligence and digital data rather than agent infrastructure, but it shows that web-derived data products can support public-market scale [Similarweb, October 2026]. If Context.dev were to become a core infrastructure layer for agent builders, a multi-billion-dollar outcome is plausible on a scenario basis, not a forecast. Under the "Agent infrastructure default" scenario, a business with broad developer adoption, durable API usage, and embedded distribution across AI workflows could reasonably support a valuation in the low single-digit billions (scenario, not a forecast), particularly if customer growth continues from 400-plus to 800-plus within the same year and expands into enterprise-grade workloads [Y Combinator, October 2026] [Context.dev, October 2026]. The public evidence does not yet support precision on revenue, retention, or margins, so the upside case should be read as category potential with early traction, not as a terminal value model.
Lightly corroborated -- Relies on company materials and Y Combinator for product scope and customer counts, with partial third-party corroboration from Silicon Valley Post, Runtime Wire, String, and Favtutor; the valuation framing is scenario analysis rather than an observed market fact.
Sources
Public sources
[Context.dev, October 2026] Context.dev: Web Data API for AI Agents (Scrape, Search, Crawl) | https://www.context.dev/
[Crunchbase] context.dev - Crunchbase Company Profile & Funding | https://www.crunchbase.com/organization/context-dev
[Y Combinator, October 2026] Context.dev: Live web data API for AI agents | https://www.ycombinator.com/launches/SPy-context-dev-live-web-data-api-for-ai-agents
[Runtime Wire, October 2026] Context.dev raises $9M to feed AI agents live web data | https://runtimewire.com/article/context-dev-raises-9m-web-data-api
[Contentedge, March 2026] Brand.dev Rebrands to Context.dev | https://www.contentedge.com/
[Yahia Bakour, LinkedIn, retrieved 2026] Yahia Bakour | LinkedIn | https://www.linkedin.com/in/yahiabakour/
[Acquire.com Blog, February 2026] How Yahia Bakour Sold Stock Alarm for $500,000 | https://acquire.com/blog/
[RocketReach] Context.dev (YC S26) | RocketReach | https://rocketreach.co/
[Silicon Valley Post, October 2026] Context.dev Raises $9M to Give AI Agents Access to the Web | https://svpost.com/articles/context-dev-9m/
[String, 2026] Best Web Scraping APIs in 2026 | https://www.string.com/
[favtutor.com, 2026] Best Web Scraping APIs | https://favtutor.com/
[Context.dev, May 2026] We’re now backed by Y Combinator | https://www.context.dev/blog/were-now-backed-by-ycombinator
[Composio] Context.dev Integration | https://composio.dev/
[Context.dev] Context.dev: the web data API | https://docs.context.dev/introduction
Articles about Context.dev
- Context.dev's API Feeds Live Web Data to 800 AI Agents — The Y Combinator alum, led by a repeat founder, raised from General Catalyst and Paul Graham to consolidate a fragmented developer workflow.