Reflections from the Co-founder of Bitlayer: After Helping 600 People Find Jobs in 15 Years, I Decided to Stop Doing Recruitment

marsbitPubblicato 2026-01-28Pubblicato ultima volta 2026-01-28

Introduzione

After 15 years of helping over 600 people find jobs, Bitlayer co-founder Kevin He reflects on his journey from being a passionate recruiter to pivoting toward a new vision: AI-powered "digital co-founders." He initially launched COCO AI in mid-2025 to optimize recruitment using AI, but soon realized traditional hiring was becoming less relevant. Companies were hiring less, and skill gaps persisted despite matching tools. A turning point came when his team built zylos, an AI system that handled tasks like social media, coding, and data processing—enabling one person to do the work of a team. This, combined with the rise of digital employees (like Claude Code and the success of Manus, acquired by Meta), shifted his perspective. He now believes the future lies in empowering individuals with organizational-level capabilities through AI, rather than optimizing job matching. COCO 2.0 is now focused on building “digital co-founders”—AI agents that help people start and run businesses alone, without needing to hire or be hired. The essay ends with an open question: In an era where AI augments individual potential, can one person become a company?

Author: Kevin He, Co-founder of Bitlayer

1. A Programmer and "Recruitment" for 15 Years

I am Kevin, an AI × Crypto entrepreneur, co-founder of Bitlayer, previously at Huobi, and a Peking University graduate.

From programmer to technical management, and then to entrepreneur, I have changed industry tracks several times over 15 years, always passionately engaged in recruitment-related work.

Over these 15 years, I have helped over 600 people successfully secure jobs. Some were formal recruitments, but more often, it was helping friends—finding jobs, introducing employees.

There are two stories I always keep in mind.

The first is helping a 35+ brother find a job. He has two children and was under great pressure. The moment he told me he received an offer, I was even more excited than he was.

The second is helping a startup team quickly build a R&D team. From zero to operational, it took less than two months. Watching the product launch, I felt I had participated in that creation.

This is why I love this work: achieving resource matching and genuinely helping people.

But recently, I started rethinking the premise of this work.

2. COCO 1.0: From Passion to Doubt

In June 2025, with support from various parties, I formed a small team and started incubating COCO AI.

The initial idea was simple: use AI to realize career advisors and career intermediaries. Make recruitment more efficient, help more people find good jobs.

The product evolved: resume optimization → AI job matching → AI corporate recruitment.

We created two products:

Job seeker end: job.coco.xyz
Enterprise end: company.coco.xyz

We invested significantly in technology, and the product was built.

But the Go-To-Market (GTM) wasn't as smooth as imagined.

I started reviewing and discovered several realities:

Companies' recruitment needs are decreasing: Due to the economic environment and improved organizational efficiency, many companies are no longer hiring on a large scale as before.
There is a gap between job seekers' skills and company requirements: AI can help with matching, but the gap itself won't disappear.
Job seekers' skills themselves need updating: Many people's abilities have fallen behind in the rapidly changing market.

I started doubting myself.

Was it a product problem? A market problem? Or—was the direction itself problematic?

I didn't have an answer at the time.

3. The Great Changes in the External World

The second half of 2025, the external world changed.

Claude Code began to be widely applied across various industries. Not just for writing code, but truly capable of completing complex tasks. More and more people around me started discussing a term: digital employees.

In December, we internally built a system called zylos, attempting to have AI handle daily work.

Tasks included:

Daily data reporting
Social media management
Code writing and testing
Data processing
Etc.

The results shocked me.

Tasks that previously required multiple people of different roles to complete could now be done just by assigning the task.

A specific example: social media operations.

Before, it required a dedicated person spending an hour daily: finding material, writing copy, formatting, publishing, replying to comments.

Now? Tell zylos, "Post one at 12:30 PM daily, content around XX theme, confirm with me before posting," and it's done in minutes.

From needing a dedicated person to needing none, from an hour daily to minutes.

The feeling was complex. Shock, excitement, but also a tinge of unease—what did this mean for the business I was building?

Around the same time, another piece of news arrived: Manus was acquired by Meta for billions of dollars.

Manus's annual revenue had already exceeded $100 million. This wasn't a concept, not a demo, but real commercial success.

I listened to Jiyichao's final podcast interview before the sale. One sentence he said impressed me deeply:

Digital employees aren't about replacing people, but enabling one person to do the work of a team.

This statement completely matched our experience testing zylos.

4. Cognitive Update

"One-Person Company" is not a new concept.

I had heard of it before and thought it made sense. But it always felt distant, too idealistic.

Until I saw zylos handle a large volume of tasks with my own eyes.

It didn't replace one specific person; it enabled me, as one person, to possess the capabilities of a small team.

In that moment, I decided to pivot.

I rethought what I had done over these 15 years.

Old cognition was: Talent is a scarce resource; helping companies find talent has value.
New cognition is: Capability enhancement is more important than talent matching. Helping individuals acquire organizational-level capabilities holds greater value.

This isn't negating the past 15 years. Those moments of helping people find jobs were still real and meaningful.

But I saw a new possibility:

Old paradigm: Individuals依附于(attach to) organizations; organizations acquire capabilities through hiring.
New paradigm: Individuals acquire organizational-level capabilities through AI, becoming independent units of creation.

The future trend is the era of one-person companies, not being employed by others.

5. COCO 2.0: Digital Co-founder

After figuring this out, the direction of COCO 2.0 changed.

Not helping you hire people, but helping you not need to hire people.

Not helping you find a job, but helping you create work.

What we are building is a "Digital Co-founder"—your AI联合创始人(Co-founder).

It has several core features:

Self-learning evolution: Not a fixed tool, but a partner that can continuously evolve based on your needs.
Secure and controllable: Your data, your business logic, all under your control.
Truly capable of doing things: Not a chat bot, but a digital employee that can complete complex tasks.

The specific product form is still being refined. But the direction is clear.

6. An Ongoing Story

As I write this article, the transition is still in progress.

Startups rarely publicize their pivots. Because of uncertainty, fear of being questioned, because there's no success yet to prove it.

But I feel this process itself is worth sharing.

When technology enables individuals to possess organizational capabilities, how will employment relationships evolve?

This is the question I am pondering, and the direction I am practicing.

If you are also thinking about these things:

If you didn't need to hire, what could you do?
If you didn't need to find a job, what would you want to do?
Can one person become a company?

This is an ongoing story.

I don't know how it will end either. But I know: Don't do things that go against the direction of the times.

15 years helping people find jobs. Next, helping people become companies.

Related reading: Interview with Bitlayer Co-founder Charlie: The Bull Market Ambition of Institutional-Grade Bitcoin Financial Infrastructure

Domande pertinenti

QWhat was the author's initial motivation for starting COCO AI, and what did they learn from its first iteration?

AThe author's initial motivation was to use AI to make recruitment more efficient, helping more people find good jobs. From the first iteration (COCO 1.0), they learned that the market had fundamental challenges: companies' hiring demands were decreasing, there was a significant skills gap between job seekers and company requirements, and many job seekers' skills were outdated.

QWhat key external event in late 2025 fundamentally changed the author's perspective on the future of work?

AThe widespread application of Claude Code across various industries and the rise of 'digital employees' was the key external event. This was further solidified by the multi-billion dollar acquisition of Manus by Meta, proving the commercial success of the digital employee model.

QWhat was the pivotal internal experiment the author's team conducted, and what was its surprising result?

AThe team built an internal system called 'zylos' to handle daily work tasks like data reporting, social media management, and coding. The surprising result was that tasks which previously required multiple people from different roles could now be accomplished simply by assigning the task to the AI, demonstrating that one person could possess the capabilities of a small team.

QHow did the author's core belief about the value of recruitment shift after their 'cognitive update'?

AThe author's core belief shifted from the old认知 (cognition) that 'talent is a scarce resource, and helping companies find talent is valuable' to the new认知 that 'capability enhancement is more important than talent matching. Helping individuals acquire organizational-level capabilities holds greater value.'

QWhat is the new mission and product direction for COCO 2.0, as described by the author?

AThe new mission for COCO 2.0 is to help people not need to hire others and to help people create work instead of find it. The product direction is to build a 'Digital Co-Founder'—an AI partner that is self-learning, secure, controllable, and capable of performing complex tasks to empower individuals to become their own companies.

Letture associate

Zcash Bug Could Have Minted Unlimited ZEC Undetected

A critical vulnerability in Zcash's Orchard shielded pool, discovered by researcher Taylor Hornby on May 29, 2026, could have allowed an attacker to create an unlimited amount of undetectable counterfeit ZEC. The flaw, involving an under-constrained element in the Orchard circuit, existed from the pool's 2022 activation until an emergency fix was deployed by June 2, 2026. Hornby identified the bug using AI-assisted auditing tools and confirmed its exploitability in a test environment. Due to Orchard's privacy features, which hide transaction amounts and history, there is no cryptographic way to prove whether the vulnerability was exploited before the fix. While Shielded Labs assesses prior exploitation as unlikely, this uncertainty has sparked a debate on proving supply integrity in privacy-preserving systems. In response, Shielded Labs and other developers are exploring a network upgrade, potentially involving a new shielded pool and formal verification of the circuit rules to prevent future vulnerabilities and allow verification of the ZEC supply's integrity. ZEC's price fell nearly 45% following the disclosure.

bitcoinist1 min fa

Zcash Bug Could Have Minted Unlimited ZEC Undetected

bitcoinist1 min fa

Silicon Valley 'Startup Guru' Steve Hoffman: Web3 + AI Could Be a Trap

Silicon Valley investor and "Godfather of Startups" Steve Hoffman warns that combining Web3 with AI is likely a trap, not a promising venture. In an interview, Hoffman argues that while AI is a foundational technology touching all industries, Web3 adds complexity, friction, and regulatory risk without solving mainstream consumer or business needs. He advises founders to focus on deep, specialized applications where startups can out-iterate giants, rather than on generic features easily replicated by large tech companies. Hoffman observes that Silicon Valley will lead foundational AI research, while China excels at rapid, large-scale application and commercialization, particularly in robotics. He stresses that AI-driven autonomous agents capable of collaborative, multi-step tasks are 2-4 years away, which will cause significant job displacement. The solution is not to slow AI but to redesign business models around human-AI collaboration and reform social systems like education and retraining. For startups, Hoffman recommends focusing on vertical, expertise-heavy domains to build defensibility. He sees major opportunities in AI fraud detection and cybersecurity. Key founder mindsets include systemic thinking over feature-focus, relentless customer centricity, building adaptive teams, and deeply understanding AI's capabilities and limits. Hoffman is also leading a non-profit initiative to establish university centers aimed at training future leaders in responsible, human-value-aligned AI innovation.

marsbit1 h fa

Silicon Valley 'Startup Guru' Steve Hoffman: Web3 + AI Could Be a Trap

marsbit1 h fa

Token Inefficient, Economy Tokenless

The article "Tokens Aren't Economical, Economics Aren't Tokenized" analyzes a pivotal shift in the AI industry from a technology-driven narrative to one dominated by capital efficiency. It highlights two concurrent trends: a severe capital shortage due to the exorbitant and recurring costs of compute (e.g., OpenAI's high burn rate) and a wave of corporate spin-offs where major tech companies are separating their AI units (like Kuaishou's Kling and Baidu's Kunlunxin). The core argument is that AI's "anti-internet" business model, where user growth increases costs rather than profits, has created a disconnect between high valuations and actual cash flow. Spin-offs address this by allowing AI assets to be valued independently. Within a parent company, they are seen as cost centers, but as standalone entities, they are priced based on their growth potential and scarcity in the primary market, leading to massive valuation premiums (e.g., Kling's estimated value tripling post-spin-off). The industry is at an inflection point, moving from "model worship" to "value realization." The competition is evolving from a pure compute (GPU) race to a broader focus on systemic efficiency and full-stack engineering (involving CPUs and orchestration) to achieve viable commercialization. The year 2026 is framed as a critical moment where the industry must definitively answer how to economically translate AI capability into tangible business value, reshaping the sector's future power structure.

marsbit1 h fa

marsbit1 h fa

Crossing the 'Memory Wall': The Wafer-Level Revolution and Computing Power Routes in the AI Inference Era

In 2026, a historic shift occurred in AI as major cloud providers' inference spending surpassed training spending for the first time, signaling a move from "building large models" to "using large models." This shifts the core challenge from computing power to the "memory wall"—the bottleneck of data movement (model weights, activations, KV Cache) between external DRAM and processors, where energy and latency from data transfer far exceed computation itself. Companies like Nvidia face GPU idle time due to bandwidth limits. In contrast, Cerebras Systems adopts a radical "wafer-scale" approach with its Wafer-Scale Engine (WSE). Instead of cutting a silicon wafer into many chips, Cerebras uses almost the entire wafer as one massive chip (WSE-3). This design provides 44GB of on-chip SRAM, delivering memory bandwidth thousands of times higher than traditional HBM (e.g., 21 PB/s vs. Nvidia B200). For LLM inference, weights are streamed layer-by-layer from external MemoryX storage to the chip, avoiding HBM bottlenecks. This results in token generation speeds 1.5–5 times faster than Nvidia's B200 in some models and significant advantages in first-token latency and long-context tasks. Additionally, Cerebras's architecture offers much lower interconnect power consumption (0.15 pJ/bit vs. GPU's ~10 pJ/bit). However, Cerebras faces challenges: SRAM scaling has slowed with advanced nodes, limiting future capacity gains; the chip requires specialized liquid cooling and custom software stacks; and its external I/O bandwidth (150 GB/s) is low compared to NVLink, hindering multi-system scaling for very large models. Competition is intensifying. Major players are pursuing three paths: 1) Developing proprietary inference ASICs (e.g., Google TPU, Microsoft Maia), 2) Leveraging advanced packaging (e.g., TSMC's SoW) to democratize wafer-scale-like integration, potentially eroding Cerebras's process advantage within a few years, and 3) Exploring optical interconnects for ultimate bandwidth. Commercially, Cerebras is transitioning from a hardware vendor to a service provider, facing the immense challenge of building high-power, specialized data centers to meet large contracts (e.g., 250MW/year from 2026–2028). In conclusion, the AI inference era presents a fundamental architectural trade-off. Cerebras opts for extreme physical optimization for low-latency, single-task performance, while Nvidia prioritizes versatility and massive cluster throughput. The path forward remains uncertain, with technology and business models still evolving in the race toward advanced AI.

marsbit1 h fa

Crossing the 'Memory Wall': The Wafer-Level Revolution and Computing Power Routes in the AI Inference Era

marsbit1 h fa

Has Bitcoin's 'Rebound Ended', Officially Entering the Late Bear Market Phase?

**Title: Has Bitcoin's Rebound Ended, Entering the Late Bear Market Phase?** **Summary:** Bitcoin's price has declined by 13% this week, signaling a potential return to late-stage bear market conditions. The price fell to around $67k, positioned between the Realized Price and Realized Cap Weighted Average. For the first time since early 2022, the Short-Term Holder cost basis has dropped below this key average, confirming a hallmark of late-cycle bear markets. Profitability metrics have collapsed sharply. The 7-day average of the Realized Profit/Loss ratio plummeted from a local high of 3.16 to 0.29, mirroring the February panic sell-off. Critically, the 90-day average never breached the threshold of 2, indicating the recent rally to $82k was a bear market bounce, not a structural shift. Realized losses surged to $1.35 billion daily, with $770 million coming from Long-Term Holders selling at a loss. This accelerating redistribution of supply from weak to strong hands is a necessary but ongoing process for a market bottom. The rally stalled almost precisely at the aggregate cost basis (~$83k) of US spot Bitcoin ETF investors, turning that level into strong resistance and leaving the average ETF holder underwater again. Spot market flows have turned decisively negative, showing sellers are dominating order books despite the price drop. While a significant futures long liquidation event cleared over $400 million in leverage, providing a potential reset, sustained spot demand is yet to materialize. Options markets continue to price in higher future volatility (Implied Volatility) than recent price action (Realized Volatility) has shown, with a persistent skew towards put options, indicating ongoing demand for downside protection. In conclusion, multiple metrics point to a fragile market structure. Resistance at the ETF cost basis, accelerating realized losses, dominant spot selling, and cautious options pricing all suggest the bear market trend persists. A sustainable recovery likely requires a resurgence of spot demand, ETF holders returning to profit, and a clear reduction in selling pressure.

marsbit1 h fa

Has Bitcoin's 'Rebound Ended', Officially Entering the Late Bear Market Phase?

marsbit1 h fa

Trading

Spot

Futures

Articoli Popolari

Come comprare PEOPLE

Benvenuto in HTX.com! Abbiamo reso l'acquisto di ConstitutionDAO (PEOPLE) semplice e conveniente. Segui la nostra guida passo passo per intraprendere il tuo viaggio nel mondo delle criptovalute.Step 1: Crea il tuo Account HTXUsa la tua email o numero di telefono per registrarti il tuo account gratuito su HTX. Vivi un'esperienza facile e sblocca tutte le funzionalità,Crea il mio accountStep 2: Vai in Acquista crypto e seleziona il tuo metodo di pagamentoCarta di credito/debito: utilizza la tua Visa o Mastercard per acquistare immediatamente ConstitutionDAOPEOPLE.Bilancio: Usa i fondi dal bilancio del tuo account HTX per fare trading senza problemi.Terze parti: abbiamo aggiunto metodi di pagamento molto utilizzati come Google Pay e Apple Pay per maggiore comodità.P2P: Fai trading direttamente con altri utenti HTX.Over-the-Counter (OTC): Offriamo servizi su misura e tassi di cambio competitivi per i trader.Step 3: Conserva ConstitutionDAO (PEOPLE)Dopo aver acquistato ConstitutionDAO (PEOPLE), conserva nel tuo account HTX. In alternativa, puoi inviare tramite trasferimento blockchain o scambiare per altre criptovalute.Step 4: Scambia ConstitutionDAO (PEOPLE)Scambia facilmente ConstitutionDAO (PEOPLE) nel mercato spot di HTX. Accedi al tuo account, seleziona la tua coppia di trading, esegui le tue operazioni e monitora in tempo reale. Offriamo un'esperienza user-friendly sia per chi ha appena iniziato che per i trader più esperti.

441 Totale visualizzazioniPubblicato il 2024.12.12Aggiornato il 2026.06.02

Discussioni

Benvenuto nella Community HTX. Qui puoi rimanere informato sugli ultimi sviluppi della piattaforma e accedere ad approfondimenti esperti sul mercato. Le opinioni degli utenti sul prezzo di PEOPLE PEOPLE sono presentate come di seguito.