Insiders: DeepSeek is forming a Harness team to compete with Claude Code
Author｜Wang Bo, Jiazi Light Year
"Jiazi Light Year" has learned from sources close to DeepSeek that the company is organizing a new Harness team focused on code intelligent agent products, internally benchmarking against Claude Code from Anthropic.
DeepSeek's senior researcher Chen Deli recently confirmed this on social media, stating, "DeepSeek is organizing a new Harness team to work on products and research in the Harness direction," and bluntly said, "In simple terms, it is benchmarking Claude Code to create DeepSeek Code Harness."
This is not an ordinary recruitment.
The job postings indicate that DeepSeek is opening two key positions: Harness Product Manager and Harness R&D Engineer, with the work location currently limited to Beijing. DeepSeek's office in Beijing is located in the Haidian District Rongke Information Center, very close to Peking University and Tsinghua University. Officially, it is located in the "Centennial Beijing-Zhang AI Innovation Belt," while informally, it is also in the recently popular "Wang Huiwen Area."
Core Definition: Model + Harness = Agent
In the job description, a core formula is prominently displayed:
Model + Harness = Agent.
This phrase can almost be seen as DeepSeek's internal definition of the productization path for the next stage: the model itself is merely the foundation of the Agent, while context management, tool invocation, task planning, file reading and writing, code modification, terminal execution, feedback collection, and evaluation loops outside the model are the key parts that allow the Agent to truly enter the workflow.
The job posting further states, "We are transforming DeepSeek's cutting-edge model capabilities into leading Agent products. All work outside of the model itself falls under the scope of Harness." Additionally, this position will participate in the entire process of "DeepSeek desktop Agent product" and "define DeepSeek's understanding of Harness."
"Jiazi Light Year" analyzes that DeepSeek is not simply trying to create a code assistant plugin, but is filling in the intermediate layer that connects the model to the real workflow.
Over the past year, the industry has proven that strong coding capabilities do not equate to developers actually using them; a model that can write code does not mean it can continuously complete an engineering task.
What truly changes the way developers work is not a standalone Claude model, but Claude Code; not a standalone GPT model, but Codex; not a code answer in a chat box, but an engineering intelligent agent that can enter the terminal, understand projects, read and write files, run commands, fix errors, manage Git, and invoke tools.
DeepSeek's greatest strength in the past was the model. Now, it is starting to add that layer of "hands" above the model.
I. Why DeepSeek Emphasizes Harness
In the context of traditional AI products, "code assistant" usually refers to two types of products: one is a completion plugin in an IDE, and the other is code Q&A in a chat box.
However, the term that repeatedly appears in DeepSeek's recruitment is not Code Assistant, but Harness.
Harness originally refers to "test harness" or "execution framework" in the engineering context, and in the Agent context, it is closer to a system that truly enables the model to take action. The model is responsible for understanding, reasoning, and generating, while Harness is responsible for integrating these capabilities into the real environment.
The job description mentions that this role needs to plan the DeepSeek Harness product roadmap, connect researchers, engineers, open-source communities, and end-users, and communicate deeply with researchers from the model training team to achieve the co-evolution of the model and Harness.
This statement is crucial.
It indicates that what DeepSeek wants to do is not just to wrap the existing model in a shell, but to make the Agent product itself a part of the model's evolution. In the past, the common product logic for large model companies was: the research team first trains a model, and then the product team creates applications based on the model's capabilities. However, in the Agent era, this order is being disrupted. Products are no longer just an outlet for model capabilities, but a training ground for those capabilities.
A code Agent failing in a real project may not be due to product interaction issues, but rather the model's incorrect compression of long contexts; it may not be a problem with the tool invocation chain, but rather the model's unstable strategy for task decomposition; or it may not be a lack of coding ability, but rather its failure to maintain a continuous understanding of engineering constraints, testing feedback, and user intent.
Therefore, the value of the Harness team is not just to "make products," but to turn real development tasks into a feedback source for the model's continuous evolution.
II. Why DeepSeek Must Fill the Code Harness Gap
DeepSeek has long bet on coding capabilities. From DeepSeek-Coder to DeepSeek-Coder-V2, DeepSeek's investment in code models has continued to increase, supporting improvements in language, context length, and complex task capabilities. Its issue is not the lack of coding ability, but rather that this capability has largely remained at the model level and has not yet become a high-frequency product in developers' daily workflows.
The popularity of Claude Code proves one thing: The competition in AI coding is shifting from competition in model capabilities to competition for entry points into developers' workflows.
This is also a lesson that DeepSeek must learn now. More subtly, before DeepSeek officially took action, the developer community had already created a version of "DeepSeek's Claude Code."
An open-source project named DeepSeek-TUI previously became popular in the developer community. It is a coding agent running in the terminal that can read and write files, execute shell commands, search the web, manage Git, and coordinate sub-agents through a TUI interface.
The rise of DeepSeek-TUI indicates two issues:
Basic Mentality Maturity: The DeepSeek model has already established a foundation in developers' minds for becoming a code Agent. Otherwise, the community would not have naturally grown around it to create Claude Code-like products.
Official Level Absence: What DeepSeek lacks is not model attention, but official Harness.
In the eyes of developers, the appeal of DeepSeek-TUI is straightforward: low cost, available domestically, long context, and relatively low deployment barriers. Many domestic developers do not want to use Claude Code, but are limited by price, access stability, account systems, and corporate compliance.
However, community projects also have inherent boundaries:
No matter how active a third-party open-source project is, it is difficult to truly grasp the evolutionary rhythm of the model's internal capabilities;
It can adapt around the API, but cannot reverse-decide how the model is trained;
It can optimize prompts, toolchains, and interactions, but it is challenging to systematically inject massive real task feedback into model improvements.
The significance of official Harness lies precisely here.
DeepSeek's own Code Harness has several advantages that community projects do not possess: collaboration with the model team, interface design authority, training data feedback loops, internal real task scenarios, and long-term operational capabilities for the developer ecosystem.
The open-source community has already paved the way: developers indeed need a DeepSeek version of Claude Code. Now, DeepSeek needs to reclaim this path and turn it into its core product.
The fact that DeepSeek is now starting to recruit means it is finally ready to step into the arena itself.
Chen Deli mentioned at the 2025 World Internet Conference in Wuzhen last November: "One of our core advantages is long-termism, insisting on making breakthroughs in cutting-edge intelligence as the main line. In this process, we have also abandoned many side projects and do not engage in those short-term, quick-return side endeavors."
After the model wars, the real Agent wars have begun. What DeepSeek needs to fill this time is the most critical layer between the model and action—Harness.
DeepSeek is equipping its model with a pair of hands.
You may also like
CFTC Reportedly Plans New Prediction Market Rules Focused on Manipulation Risk and Public Interest Review
The CFTC is reportedly preparing new prediction market rules focused on manipulation risk, public interest review, and retail trader protections.
Meet the new WEEX trial fund—your gateway to greater profits
WEEX Labs Lands at Dutch Blockchain Week: A Disruptive Crypto × AI Conversation Sets Sail in Amsterdam
SK Hynix Reportedly Plans U.S. ADR Listing as Early as August, With SEC Approval Possible in Late June
SK Hynix may pursue a U.S. ADR listing as early as August, with SEC approval reportedly possible in late June amid strong AI chip supply chain demand.
SpaceX vs Tesla vs xAI: Which Elon Musk Trade Has the Biggest Upside in 2026?
OpenAI Reveals It Has Confidentially Submitted an S-1 to the SEC, Keeping the Door Open for a Future IPO
On June 9, according to an OpenAI announcement, the company recently confidentially submitted a draft S-1 registration statement to the U.S. Securities and Exchange Commission (SEC), beginning the preliminary compliance process for a potential initial public offering. OpenAI said it chose to disclose this proactively because it expected the news might leak; however, the company has not yet set a specific listing timeline, and related arrangements may still take some time.
Latest research from 13 top universities including Cornell University: The current state, challenges, and misconceptions of the fusion of Crypto and AI
Deconstructing Anthropic: The Best AI Company, Possibly Also a Type of Organizational Invention
Apollo and Blackstone Reportedly Back $35 Billion Anthropic Chip Financing as Deal Details Remain Unclear
On June 9, according to currently available news alerts, Apollo and Blackstone Group participated in a $35 billion financing for an Anthropic “chip project.” Based on the original wording of the report, the funding has already been raised, but public information remains limited. The financing structure, use of proceeds, project entity, and whether Apollo and Blackstone participated through equity, debt, or project financing have not yet been disclosed.
Humanity Protocol Security Incident Escalates: More Than $31 Million Stolen From Related Addresses as Attacker Continues Selling H for ETH
On June 9, according to monitoring by Onchain Lens, more than $31 million has been stolen from addresses linked to Humanity Protocol, and the attack is still ongoing, with the hacker continuously swapping H tokens for ETH. Project founder Terence Kwok later confirmed the security incident on X, saying the issue involved a private key leak.
Bloomberg: As Bitcoin Weakens, Stablecoins and RWA Continue to Drive Expansion in Crypto Businesses
In June, Bloomberg reported that despite Bitcoin falling below $60,000 last week, wiping out about $235 billion in market value within seven days, and dropping close to 50% from last year’s peak, some core businesses in the crypto industry are still expanding, mainly in stablecoins, real-world asset tokenization (RWA), payments, and infrastructure. The report also noted that overall altcoin activity has contracted significantly: altcoin market capitalization has fallen from a peak of about $431 billion in November 2021 to around $170 billion, and among the tens of millions of tokens issued in recent years, fewer than 1,700 still maintain meaningful trading activity.
Galaxy Deep Research Report: How Hyperliquid's HIP-4 Upgrade Changes the Landscape of Prediction Markets?
Binance Research: RWA Market Expected to Expand Nearly 6x from Early 2025, with Public Equities and Onchain Payments Heating Up Together
In June, Binance Research said in its monthly market report that the real-world asset (RWA) market is expected to grow by about 589% from the beginning of 2025. Bond- and money market fund-related RWA expanded by about $6.5 billion, up 83% year over year, while publicly traded equity RWAs grew by about 422%. The report also noted that monthly crypto debit card transaction volume exceeded $747 million in May, up 48.6% year to date.
Japan to Assess a Framework for Yen Stablecoins and Crypto ETFs as Asia’s Compliant Payments Narrative Heats Up
Recently, according to the original report, Japan is considering the launch of yen stablecoins and cryptocurrency ETFs. Public information remains limited at this stage, and there is still no complete policy text, regulatory draft, or clear implementation timeline, so this is better characterized as a “policy discussion” rather than formal implementation. The original wording also noted that advancing stablecoin regulation in Asia is driving XRP usage and supporting growth in the XRPL ecosystem. However, based on currently available public information, there is not enough evidence to directly establish a clear causal relationship between this round of discussion in Japan and XRP or XRPL.
ZachXBT: Humanity private key leak and abnormal surge in H token should be viewed separately
On June 9, according to related disclosures, on-chain investigator ZachXBT posted an update on Humanity’s roughly $31 million security incident, saying that after further analyzing fund flows, he currently tends to believe the project team was not involved in an “inside job” or a self-staged attack. According to him, the official explanation about the private key leak was broadly accurate, but before the token unlock, the price of H had been artificially pushed higher, and the hacker later took advantage of that market environment; therefore, the private key leak and the earlier abnormal price pumping should be regarded as two separate and independent events. This reframing has shifted the market’s understanding of the nature of the incident. Earlier discussion around Humanity had focused on whether the team directly participated in the attack or used the security incident to cover up internal operations. ZachXBT’s latest remarks shift the focus from “whether it was self-theft” to “whether there were pre-unlock market structure issues.” He also questioned whether the team may have.
Morning Report | OpenAI has submitted an S-1 registration statement draft to the U.S. SEC; Morpho completes $175 million financing
Morning Report | BitMine increased its holdings by 126,971 ETH last week; trader Eugene announced his exit from the crypto market
Wang Chuan: How can one not feel anxious after the neighbor Old Wang made thirty times profit by investing in storage stocks? (Seven) - A quarter-century cycle
CFTC Reportedly Plans New Prediction Market Rules Focused on Manipulation Risk and Public Interest Review
The CFTC is reportedly preparing new prediction market rules focused on manipulation risk, public interest review, and retail trader protections.
Meet the new WEEX trial fund—your gateway to greater profits
WEEX Labs Lands at Dutch Blockchain Week: A Disruptive Crypto × AI Conversation Sets Sail in Amsterdam
SK Hynix Reportedly Plans U.S. ADR Listing as Early as August, With SEC Approval Possible in Late June
SK Hynix may pursue a U.S. ADR listing as early as August, with SEC approval reportedly possible in late June amid strong AI chip supply chain demand.
SpaceX vs Tesla vs xAI: Which Elon Musk Trade Has the Biggest Upside in 2026?
OpenAI Reveals It Has Confidentially Submitted an S-1 to the SEC, Keeping the Door Open for a Future IPO
On June 9, according to an OpenAI announcement, the company recently confidentially submitted a draft S-1 registration statement to the U.S. Securities and Exchange Commission (SEC), beginning the preliminary compliance process for a potential initial public offering. OpenAI said it chose to disclose this proactively because it expected the news might leak; however, the company has not yet set a specific listing timeline, and related arrangements may still take some time.
