Datacurve

Commercial

medium confidence

datacurve.ai ↗ · Status: active confirmed · Founded 2024

Datacurve is a YC W24 data vendor that says it supplies frontier labs with expert-sourced RL environments, long-horizon tasks, agent trajectories, SFT demonstrations and off-the-shelf datasets. It is coding-first but now also covers data science, cyber security, ML and research. Data comes from its Shipd bounty platform of vetted engineers, which has SWE, ML and data-science tracks. It publishes DeepSWE, an open (Apache-2.0) long-horizon coding benchmark used in the Artificial Analysis Coding Agent Index.

Key facts

Headquarters

San Francisco, USAconfirmed cite

Headcount band

11-50reported cite

Total raised

$17.7Mconfirmed cite

Last round

Series A, $15M, October 2025confirmed cite

SOC 2

unknown

What they sell

mixedconfirmed cite

Open source

no (company data products are commercial/proprietary; the DeepSWE benchmark and harness are open-source under Apache-2.0)confirmed cite

Deployment

managed-hosted (data delivered as a service; private model endpoints spun up for RLHF traces)reported cite

What's new

8 sourced updates · last 6 months

5 Oct 2026product Product line now spans multiple RL-environment domains; Shipd lists SWE, ML and data-science tracks
The products page lists six offerings: reinforcement learning environments (software engineering, data science, cyber security, machine learning, research), long-horizon tasks, OTS datasets, benchmarks and evals, agent trajectories, and supervised fine-tuning. Shipd.ai lists three bounty tracks: Mars (software engineering), Eris (machine learning) and Monte (data science). The exact change dates are unknown; both pages were observed on 2026-10-05.
Why it matters: Datacurve now presents itself as a multi-domain vendor of RL environments and agent data, and its contributor supply now covers ML and data-science work as well as coding.
3 Sep 2026benchmark DeepSWE leaderboard updated with new frontier models (latest: GPT-6 Astra, Gemini 3.8 Flash)
Between 2026-06-21 and 2026-09-03 Datacurve added results for GLM 5.2, Claude Sonnet 5, GPT-5.6 Sol/Terra/Luna, Muse Spark 1.1/1.2, Grok 4.5/4.6, Kimi K3, Gemini 3.6/3.7/3.8 Flash, Claude Opus 5, Qwen 3.8 Max, DeepSeek v4 Flash/Pro and GPT-6 Astra. It also corrected pricing and re-ran Gemini models after a token-counting issue.
Why it matters: The leaderboard is actively maintained, so it is a current reference rather than a one-off release.
26 Aug 2026open-source DeepSWE repo is Apache-2.0 licensed and Harbor-format, about 1.8k stars
The GitHub API reports Apache License 2.0, a last push on 2026-08-26 and 1,793 stars (652 in the 2026-06-07 baseline, when no license was set). Tasks use the Harbor format. Since v1.1 they run through Pier, a Harbor-compatible sandbox framework with per-agent network allowlists. The date is the last push; the exact date the license was added is not known.
Why it matters: An open license and a standard Harbor/Docker task format make the benchmark easy for labs and data-ops teams to reproduce and integrate.
29 Jun 2026research Datacurve founding engineer James Shi presents DeepSWE at AI Engineer World's Fair 2026
Talk: 'DeepSWE: A Contamination-Resistant Coding Benchmark' (17:34). The World's Fair ran June 29-July 2, 2026 in San Francisco. The date shown is the event's first day; the exact talk day is not confirmed. A StartupHub write-up of the talk is dated 2026-07-26.
Why it matters: A public talk on how the benchmark was built, useful to researchers judging its contamination resistance.
19 Jun 2026benchmark DeepLearning.AI's The Batch covers DeepSWE as a harder successor to SWE-bench
The Batch (2026-06-19) covered DeepSWE, 'developed by Datacurve', alongside other agentic tests. It reported GPT-5.5 (xhigh) at 70%, Claude Opus 4.8 at 58% and Gemini 3 Flash at 5%.
Why it matters: Coverage in a widely read research newsletter adds to the benchmark's credibility and reach.
14 Jun 2026benchmark DeepSWE v1.1 released with isolated verification and structured test reports
Written by Wenqi Huang and Peter Jiang. The agent's committed git patch is now graded in a separate isolated container. Tests emit CTRF structured reports. Tasks start on a normal main branch rather than a detached HEAD, with future commits hidden. Flaky tests and dependency drift were fixed. The task set is the same 113 tasks. The changelog records the v1.1 release on 2026-06-15.
Why it matters: Isolated grading with per-test reports makes model comparisons more reproducible and makes partial-progress analysis possible.
12 Jun 2026benchmark Artificial Analysis replaces SWE-Bench-Pro-Hard with DeepSWE in its Coding Agent Index
Artificial Analysis's methodology page says Coding Agent Index v1.1 (June-July 2026) introduced DeepSWE and removed SWE-Bench-Pro-Hard-AA. Version 1.5 (September 2026, current) is an equal-weight average of DeepSWE v1.1, Terminal-Bench 4.0 and SWE-Atlas-QnA. AlphaSignal covered the switch on 2026-06-12 and reported that Claude models had recovered gold solutions from .git history on 12%+ of SWE-Bench Pro runs. The date is AlphaSignal's coverage date; the index changelog gives only the month.
Why it matters: A major independent evaluator now uses Datacurve's benchmark, which gives it standing beyond Datacurve's own claims.
26 May 2026benchmark Datacurve releases DeepSWE, a long-horizon agentic coding benchmark
The launch post 'Introducing DeepSWE' is dated 2026-05-26 and written by Wenqi Huang, Charley Lee, Leonard Tng and Serena Ge. Datacurve's research index lists it as 2026-05-18, and the GitHub repo was created 2026-05-15. The benchmark has 113 tasks across 91 repositories in 5 languages (TypeScript, Go, Python, JavaScript, Rust). Reference solutions are written from scratch, average 668 added lines and touch about 7 files. Datacurve reports a 0.3% verifier false-positive rate against 8.5% for SWE-Bench Pro, and 1.1% false negatives against 24.0%. At launch GPT-5.5 scored 70% and Claude Opus 4.7 scored 54%. All of these figures are self-reported.
Why it matters: This is Datacurve's first major public research artifact. It is a contamination-resistant coding eval that separates frontier models more widely than SWE-bench, and it shows the kind of task authoring Datacurve sells.

Founding team

Serena Ge

Co-founder & CEO

Studied CS at the University of Waterloo before dropping out to found Datacurve. Worked with Cohere's CTO on LLM reasoning and coding through synthetic data generation. Forbes 30 Under 30 (per the University of Waterloo).

Profile ↗

Charley Lee

Co-founder

University of Waterloo CS. Former Google intern. AI research on multi-modal RL and browser-use agents (per his YC profile). Co-author of the DeepSWE launch post.

Profile ↗

What practitioners should know

For researchers

DeepSWE tasks are written from scratch rather than mined from merged PRs, to resist contamination. Datacurve reports verifier false positives of 0.3% against 8.5% for SWE-Bench Pro, and false negatives of 1.1% against 24.0%, measured on an audited sample. These are self-reported figures.
Graded per-rollout trials can be inspected at deepswe.datacurve.ai/data/v1.1/trials.
DeepSWE v1.1 grades the committed git patch in a fresh isolated container and emits per-test CTRF reports, so partial-progress analysis is possible.
Artificial Analysis uses DeepSWE v1.1 as one of three equally weighted components of its Coding Agent Index (v1.5, September 2026).

For program managers

Engagement is project-based custom data and environments, through contact@datacurve.ai. No named customers are published, and frontier-lab customer ties remain self-described and unverified.
Funding is unchanged since the October 2025 Series A ($15M led by Chemistry), about $17.7M in total. No new round was found as of 2026-10-05. Tracxn's $18.2M total is an unreconciled aggregator figure.
Small team: LinkedIn shows 29 employees (2026-10-05), against about 36 in the 2026-06 baseline. Tracxn reports 28 as of 2026-07-31. The careers page lists 3 open roles in San Francisco.
The offering has widened from coding data to multi-domain RL environments, long-horizon tasks, agent trajectories and OTS datasets, and Shipd now has ML and data-science tracks.

For data ops

DeepSWE tasks use the Harbor format (task.toml, instruction.md, Docker environment/, tests/ verifier, solution/) and run through Pier, a Harbor-compatible sandbox with per-agent network allowlists. Pier also supports claude-code, codex, gemini-cli and opencode agents.
The DeepSWE repo is Apache-2.0 licensed.
No public security or trust page and no SOC 2 attestation were found.

Benchmarks & research

3 published

DeepSWE leaderboard and trials browserleaderboard · 15 Jun 2026

Maintained leaderboard of pass@1 and average cost per model configuration, updated as new frontier models ship. Per-rollout graded trials are browsable at deepswe.datacurve.ai/data/v1.1/trials.

Top score 74% pass@1 (three-way tie, as of 2026-09-22)

Open ↗

DeepSWE (v1.1)benchmark · 26 May 2026

Long-horizon, contamination-resistant coding-agent benchmark: 113 original tasks written from scratch, across 91 repositories in 5 languages (TypeScript, JavaScript, Python, Go, Rust). Graded by hand-written behavioral verifiers. v1.1 (2026-06-14) grades the committed patch in an isolated container. The standard harness is mini-swe-agent. It is a component of the Artificial Analysis Coding Agent Index (v1.1 onward).

As of 2026-09-22, GPT-6 Astra, Gemini 3.8 Flash and Claude Opus 5 tie at 74% pass@1 on the leaderboard. At launch GPT-5.5 led with 70%.

Open ↗Paper ↗

DeepSWE GitHub repositorydataset · 15 May 2026

Open-source task set and harness (Apache-2.0). Tasks use the Harbor format (task.toml, instruction.md, environment/, tests/, solution/) and run through Pier, a Harbor-compatible sandbox. 1,793 stars as of 2026-10-05.

Open ↗

Products

Reinforcement Learning environments ↗

Environments for measuring and training agent capability in software engineering, data science, cyber security, machine learning and research.

Long-horizon Tasks ↗

Tasks that run over long periods and involve ambiguity, partial progress and recovery.

OTS Datasets ↗

Pre-built, off-the-shelf training datasets.

Benchmarks and Evals ↗

Custom evaluations. DeepSWE is the public example.

Agent Trajectories ↗

Full agent execution traces.

Supervised Fine-tuning data ↗

Expert demonstrations for SFT.

Shipd ↗

Datacurve-operated bounty platform where vetted engineers solve paid problems that become training data. It has three tracks: Mars (software engineering), Eris (ML competitions) and Monte (data science). This is Datacurve's contributor supply.

DeepSWE ↗

Public long-horizon coding benchmark, leaderboard and trials browser.

Scale & velocity

Current headcount

approx 29 (LinkedIn, size band 11-50) as of 2026-10-05reported cite

Headcount growth

down slightly (LinkedIn approx 36 in 2026-06 to approx 29 in 2026-10)estimated cite

Open roles

3confirmed cite

Other locations

unknown

Distributed / remote

no (San Francisco office-based; careers page lists in-office meals and commuter benefits)estimated cite

Research depth

Has researchers

yesreported cite

Researcher count

unknown

Backgrounds

Serena Ge (co-founder/CEO): worked on LLM reasoning during a co-op at Cohere; University of Waterloo CS; Forbes 30 Under 30, Charley Lee (co-founder): University of Waterloo CS; AI research backgroundreported cite

Capital

Total raised

$17.7M ($15M Series A + $2.7M seed)confirmed cite

Last round

Series A, $15M, October 2025confirmed cite

Investors

Chemistry (Mark Goldberg, lead Series A), Y Combinator, Balaji Srinivasan (seed), angel investors who are employees of DeepMind, Vercel, Anthropic and OpenAI (individuals, not the companies)reported cite

Valuation

unknown

Revenue signals

unknown (over $1M paid out in bounties to contributors, a payout figure, not revenue) cite

Security & compliance

SOC 2

unknown

Other certifications

unknown

Security page

unknown

Product

What they sell

mixedconfirmed cite

Open source

no (company data products are commercial/proprietary; the DeepSWE benchmark and harness are open-source under Apache-2.0)confirmed cite

License

Apache-2.0 (DeepSWE benchmark repo); commercial data products proprietaryconfirmed cite

Deployment model

managed-hosted (data delivered as a service; private model endpoints spun up for RLHF traces)reported cite

Maturity

GAestimated cite

Notable customers

unknown

Buyer analysis

Best fit: Frontier/foundation model labs needing expert-sourced coding SFT/RLHF data and code-execution RL environments with verifiable rewards (code execution, tight loops).

How we verified this

2026-10-05 refresh: I opened every source in the draft. The main claims hold: - DeepSWE has 113 tasks across 91 repos in 5 languages, with reference solutions averaging 668 lines. - v1.1 came out on 2026-06-14 and was written by Huang and Jiang. - The changelog runs from 2026-06-15 to 2026-09-03. - The repo is Apache-2.0 with 1,793 stars, last pushed 2026-08-26. - Artificial Analysis made DeepSWE part of its Coding Agent Index in v1.1 (June-July 2026), and it is still a component in v1.5. - The Batch (2026-06-19) reports GPT-5.5 at 70% and Opus 4.8 at 58%. - AlphaSignal (2026-06-12) covered the switch. - The products page lists the five domains. - LinkedIn shows 29 employees. - Both YC founder headshots load as single-person images. I made these corrections: - **Launch date:** changed to 2026-05-26, from the primary launch post. - **False-positive figures:** sourced to that post and labelled as self-reported. - **Trials URL:** fixed to the deepswe subdomain. - **Reasoning-effort labels:** removed from the 74% tie because I could not verify them. - **Pier:** removed the claim that Datacurve owns it. - **Shipd:** moved the source from TechCrunch, which does not mention it, to shipd.ai. - **World's Fair talk:** re-dated to the event window (June 29-July 2). I added Shipd's new ML and data-science tracks. I found no funding, M&A or status change. My web search budget ran out during this check, so the search for missed news was limited to directly fetched company pages, LinkedIn and Tracxn, none of which showed a new round. Customer and frontier-lab ties remain unverified self-claims.

Related vendors

Sources

datacurve.ai/ · 2026-10-05, Official site, active, San Francisco, (c) 2026
datacurve.ai/products · 2026-10-05, Six offerings; RL environments in 5 domains; no customers named
datacurve.ai/research · 2026-10-05, Research index lists 'Introducing DeepSWE' as 2026-05-18
datacurve.ai/careers · 2026-10-05, 3 open roles (SWE, Research Engineer, Growth/Ops), San Francisco, in-office benefits
deepswe.datacurve.ai/ · 2026-10-05, Leaderboard as of 2026-09-22: three-way tie at 74% pass@1; 113 tasks / 91 repos / 5 languages; mini-swe-agent harness
deepswe.datacurve.ai/blog/deepswe · 2026-10-05, Launch post, 2026-05-26 (Huang, Lee, Tng, Ge); FP 0.3% vs 8.5%, FN 1.1% vs 24.0%; 668 LOC; GPT-5.5 70%, Opus 4.7 54%
deepswe.datacurve.ai/blog/deepswe-v1-1 · 2026-10-05, v1.1 post, 2026-06-14, Wenqi Huang and Peter Jiang
deepswe.datacurve.ai/changelog · 2026-10-05, Changelog entries from 2026-06-15 to 2026-09-03
deepswe.datacurve.ai/data/v1.1/trials · 2026-10-05, Browser for graded per-rollout trials
github.com/datacurve-ai/deep-swe · 2026-10-05, README: Harbor format, Pier sandbox, Apache-2.0
api.github.com/repos/datacurve-ai/deep-swe · 2026-10-05, Created 2026-05-15, pushed 2026-08-26, 1,793 stars, Apache-2.0
artificialanalysis.ai/methodology/coding-agents-benchmarking · 2026-10-05, Coding Agent Index: DeepSWE introduced in v1.1 (Jun-Jul 2026); v1.5 (Sept 2026) is an equal-weight average of DeepSWE v1.1, Terminal-Bench 4.0 and SWE-Atlas-QnA
alphasignal.ai/news/datacurve-s-deepswe-catches-claude-cheating-and-resh · 2026-10-05, 2026-06-12 coverage of Artificial Analysis switching to DeepSWE (reported)
deeplearning.ai/the-batch/agentic-tests-beyond-the-bug-hunt · 2026-10-05, The Batch, 2026-06-19, DeepSWE coverage
mer.vin/2026/05/deepswe-benchmark-how-datacurve-separates-real-agentic-c · 2026-10-05, Third-party post (2026-05-28) giving the release date as 2026-05-26
ai.engineer/orgs/datacurve · 2026-10-05, AIE World's Fair 2026 talk by James Shi, founding engineer
www.ai.engineer/worldsfair · 2026-10-05, Event dates June 29-July 2, 2026, San Francisco
www.startuphub.ai/ai-news/ai-research/2026/data-curve-launches-deepswe-c · 2026-10-05, Aggregator write-up of the AIE talk, dated 2026-07-26; not used for dating
shipd.ai/ · 2026-10-05, Shipd, operated by Datacurve AI; Mars (SWE), Eris (ML), Monte (DS) tracks
www.ycombinator.com/companies/datacurve · 2026-10-05, Founder bios, profile links and headshot URLs (both photos checked: HTTP 200 image/jpeg, single-person headshots)
www.linkedin.com/company/datacurveai · 2026-10-05, 29 employees, 11-50 band; DeepSWE posts
techcrunch.com/2025/10/09/datacurve-raises-15-million-to-take-on-scaleai · 2026-10-05, Last round: $15M Series A, Oct 2025, led by Chemistry; no newer round found
www.chemistry.vc/post/staying-ahead-of-the-curve · 2026-10-05, Lead investor post naming the shipd.ai bounty platform
tracxn.com/d/companies/datacurve/__2w9YlTVQ2FbPLzVtvcqLEI1v_XXMwKgf4xoxL · 2026-10-05, Aggregator: 28 employees as of 2026-07-31; no acquisitions (reported, last resort)
sacra.com/c/datacurve/ · 2026-06-07, Funding, RLHF traces via private endpoints, repo-wide RL environments with unit tests, 14,000 vetted engineers
uwaterloo.ca/computer-science/news/cs-led-startup-secures-177m-transform · 2026-06-07, Founder backgrounds: Ge ex-Cohere, Lee ex-Google/RL research, Waterloo CS
www.menlotimes.com/post/datacurve-is-taking-on-scale-ai-building-frontie · 2026-06-07, Bounty model, complex RL environments, future expansion plans

Last updated 2026-10-05 · Every quantitative field carries a source and a confidence tag. Fields we could not source publicly are marked unknown, never estimated. See the methodology.