MIP40c3-SP15: Modify Oracles Core Unit Budget

MIP40c3-SP15: Modify Oracles Core Unit Budget


MIP40c3-SP#: 15 
Author(s): Niklas Kunkel
Contributors: N/A
Status: Accepted
Date Applied: 2021-05-12
Date Ratified: 2021-06-28

Sentence Summary

The Oracle Core Unit is requesting an annual budget of $5M.

Paragraph Summary

The Oracle Core Unit (OCU) is responsible for developing and administrating the Oracle Protocol. Oracles are a critical component of securing the Maker Protocol to ensure positions are sufficiently collateralized. The OCU is essential to ensure the Oracles can adapt rapidly to scale the Maker Protocol in a responsible manner. A key focus of the OCU is to monetize the Oracle Protocol and build out a business around it. This enables the DAO to subsidize its own Oracle expenditures, and ultimately turn Oracles from a cost-center into a profit-generating arm of MakerDAO. Real World Assets (RWA) in particular, are a highly promising vertical the business side of the Oracle is targeting.


Core Unit Name

Oracles Core Unit


On behalf of the Oracles Core Unit, I’m proposing the following budget to deliver on our mandate. This includes, but is not limited to:

  • Securing the Maker Protocol
  • Developing the Oracle Protocol to meet the constantly growing needs of the Maker Protocol
  • Deploying, Monitoring, and Administering Oracles on behalf of the Maker Protocol
  • Responsibility for all Oracle elements of collateral-onboarding
  • Transforming the Oracles into a revenue-generating service
  • Expanding the Oracles product suite

Oracles are a critical component of the Maker Protocol’s security, through which failure could potentially allow printing of unbacked Dai, and liquidation of all vaults. Many DeFi hacks can be traced back to a faulty Oracle. The budget presented below represents the resources needed to employ the exceptional talent to take on this critical responsibility.

Budget Considerations

This budget was designed with the philosophy that it’s better to ask for too much budget and not spend it, than ask for too little, and need it. At this stage of growth for the Maker Protocol, being blocked by a lack of capital would be a careless and unnecesary mistake. There is a limited time-window to execute on the opportunities at hand before the market becomes too crowded with well-capitalized competitors. Given the amount of time it will take to hire the right people and set up internal processes, it is highly unlikely that the Oracle Core Unit will spend even half of the proposed budget in its first year. Subsequent budget requests will have much better calibration with actual expenses incurred.

The following considerations have been taken into account when building the proposed budget:

  • A competitive salary in-line with industry standards
  • A competitive benefits package in-line with industry standards
  • The Core Unit will need to find and hire exceptional talent rapidly in order to scale quickly enough to meet the needs of its mandate. To this end, a large pool of capital has been allocated for recruiting, refferals, and sign-on bonuses.
  • A high travel budget, in order to fund attending crypto conferences and team offsites which are critical for remote teams to build interpersonal relationships; vital for collaboration and retention.
  • A healthy contingency in case budget estimates are significantly off

It should be noted that the gas costs for operating Oracles are not included in this budget and will be requested in a later subproposal. These costs scale linearly with the gas-price of the Ethereum Network and are significant, variable, and difficult to model.


The yearly budget request for the Oracles Core Unit is $5m. This equates to a $419k monthly expense to support the team mandate.

This budget secures a team of 15.5 full-time employees as well as covers all operational costs listed below.

At the end of the calendar year, all unused funds will be returned to the Maker Protocol. Should the Oracle Core Unit shutdown, all unused funds will be returned to the DAO immediately.

Annual Budget

Component Amount
Salary $2,575,000
Total Employee Cost (Salary + Benefits + Taxes) $3,347,500
Feed Stipends $360,000
Recruiting/Referral/Sign-On Bonus $300,000
Audits $200,000
Travel $150,000
Tech-Ops Core Unit - Production Coverage $150,000
Legal $100,000
Marketing $100,000
Deployment Costs $50,000
Tools/Services/Devices $35,000
Hardware $30,000
Operations $25,000
Contingency $238,625
Total Annual Budget $5,036,125
Total Monthly Budget $419,677

Budget Distribution

Budget Details

The current team has 4 full-time developers, 1 Team Lead, and 1 Facilitator. The team is seeking to grow this by 9.5 members to 15.5 total to meet the evergrowing needs of the Maker Protocol.

Total Employee Cost
The total cost of an employee includes salary, all benefits (including healthcare), and taxes. This has been calculated by scaling salaries by 30%. This premium is likely an overestimation, and the realized cost is expected to be lower. That being said, paying salaries to a a global remote team in a multitide of jurisdictions can be quite expensive without the proper legal infrastructure in place.

Feed Stipends
All Feeds are compensated with a Feed Stipend of 1000 Dai per month. There are currently 26 total Feeds; 12 Light Feeds, and 14 Dark Feeds. We expect to continue onboarding new Feeds in the coming year to meet the growing security requirements of the Maker Protocol as it scales.

Recruiting/Referall/Sign-On Bonus
Recruiters are going to be critical for discovering talent, and their commission for successful hires is often high, especially for more senior roles. We also want to have flexibility with offering significant sign-on bonuses to elite cant-live-without candidates. Keep in mind many of these are one-time costs.

Conferences and team offsites have been critical in the Maker Foundation’s history of fostering interpersonal relationships and generating trust. These are vital qualities in a companies culture that improve collaboration, productivity, and retention.

Conferences in particular also serve a multitude of purposes. Employees are educated about the bleeding-edge technologies in the industry, and can apply that knowledge within the Oracle Core Unit. They also are a way for the Oracle Core Unit to present seminars, giving visability to potential customers and attracting new talent. Additionally, they serve as hotspots which can be used to connect with potential clients in a much more personal (read effective) manner.

This budget would cover flights, lodging, nourishment, team-activities, and conference passes.

As the Oracle Protocol evolves, audits are an indispensible part of the QA process. In many cases, multiples audits may be required for a single release due to the sensitive role that Oracles play in securing the Maker Protocol. Depending on the code size and complexity, as well as the prestige of the auditor, audits can range from $10,000 to $50,000. During bull markets especially, auditors are booked months in advance and charge heavy premiums for fast-tracking. The Oracle Core Unit will always prioritize the security of the Maker Protocol over agility.

Tech-Ops Core Unit - Production Coverage
The Maker Foundation Tech-Ops team has in the past provided excellent coverage in monitoring production systems 24/7 365 across a multitude of timezones. This would be prohibitively expensive to staff within the Oracle Core Unit, and is best outsourced to the Tech-Ops Core Unit which will provide round the clock monitoring to a handful of Core Units running production services on behalf of the Maker Protocol.

Coverage for legal costs including entity creation, legal officer/company insurance, as well as monthly and annual financial reporting.

The marketing budget is for building a brand around the Oracle Protocol.
As we won’t have an in-house designer initially, some of the budget will be used to outsource design work. Conference sponsorships, ad campaigns with select partners, and user incentives also fall under this umbrella.

General operational overhead as well as outsourcing accounting work (potentially to another Core Unit).

Deployment Costs (Gas)
Gas costs have continued to ride as demand for interacting on the Ethereum network grows. In particular, the plethora of AMMs which become arbitraged and front-run by bots en-masse is a systemic problem that does not look to let up anytime soon. As such we expect extreme gas-costs for smart-contract deployments, the majority of which are for Oracles related to collateral-onboarding for the Maker Protocol. This budget should cover deployment and configuration for roughly 35 collateral types.

New team members will need to be provided with laptops for the duration of their employment. Licenses for essential productivity tools like Clubhouse

The Oracle Core Unit will need to host Oracle development and production infrastructure such as testing and staging environments, developer environments, relayers, ethereum nodes, APIs, data logging, and monitoring tools.

As with any new indeavor, especially in an industry innovating as quickly as ours, there are many unknowns when it comes to estimating budget requirements, as well as unexpected costs that arise from time-to-time. The contigency represents 5% of the total budget and is there to act as a safety buffer.

Budget Distribution

The budget is to be distributed on a monthly basis to the Oracle Core Unit Multisig, which is an instance of a Gnosis Safe multisig. Note that the Maker Protocol retains full administrative control of the funds in the Oracle Core Unit Multisig. To start, the multisig has 3 signers (see below) with a quorum of 2. This ensures that no rogue member is able to control the funds,


  1. @NikKunkel (Oracle Core Unit Facilitator) - 0xB4CFf22c0d5a015460Fd503a9328Bf00a24f686c
  2. @marcandu (Oracle Core Unit Engineering Team Lead) - 0xf63EfEE2A663b04cB5a09De6E3EB910d8442131D
  3. @Derek (Protocol Engineering Core Unit Facilitator) - 0xe3a76328edE8Fd61d5fA7840b878Dd69cdfD67d8

This multisig conforms to the requirement ratified by Maker Governance in MIP47.

Conditions - Continuous Operation

To ensure 3 months of continuous operation in the event of emergency shutdown or protocol issues, the Oracle Core Unit is requesting an upfront lump-sum of $800K to sit in a multisig controlled by a set of Core Unit Facilitators.


The lump sum is the equivalent of 3 months of budgeted salaries and essential team expenses.


If normal protocol operation is not possible due to emergency shutdown or other protocol failure, these funds will be used to ensure employment resources get the system back up and functional.

At the conclusion of the year, the lump sum will remain in the multisig for the following year and may be increased at that point in time due to team growth or the need for an increased runway.

Note that this multisig is NOT controlled by Maker Governance but rather exclusively by a set of Core Unit Facilitators. This is intentional to protect the funds in an event of a governance attack. Governance can indirectly claw these funds back through a consensus of Core Unit Facilitators.


  1. @NikKunkel (Oracle Core Unit Facilitator) - 0xB4CFf22c0d5a015460Fd503a9328Bf00a24f686c
  2. @Primoz (Risk Core Unit Facilitator) - 0x5d67d5B1fC7EF4bfF31967bE2D2d7b9323c1521c
  3. @Derek (Protocol Engineering Core Unit Facilitator) - 0xe3a76328edE8Fd61d5fA7840b878Dd69cdfD67d8
  4. @LongForWisdom (Gov Alpha Core Unit Facilitator) - 0x66f40F044E0e2F77bB746e3275E82e88dCBA2D69
  5. @Nadia (Growth Core Unit Facilitator) -
  6. @SebVentures (Real World Finance Core Unit Facilitator) - 0x0D61C8b6CA9669A36F351De3AE335e9689dd9C5b
  7. @juan (Sustainable Ecosystem Scaling Core Unit Facilitator) - 0xFCa6e196c2ad557E64D9397e283C2AFe57344b75

MKR Vesting

The Oracle Core Unit is proposing to apply a linear interpolation model to utilize salary to calculate the amount of MKR each individual will vest. This model has been proposed by the SES Core Unit with minor modifications. These modifications include a higher floor for intermediate and junior contributors as well as a longer vesting period to reward loyal contributors and mitigate brain-drain.


[Initial Annual FTE Incentive Value in USD] = MAX( 
    2.45 * [Annual FTE Gross Salary] -275,000 USD; 
    [Annual FTE Gross Salary] * 0.35
[Initial Annual FTE Incentive (MKR)] = 

    [Initial Annual FTE Incentive Value in USD] /
    [180-Average Price of MKR at the first date of employment]

Vesting Details

Property Value
MKR/USD lock-in Price (New) Trailing 6 month average
MKR/USD lock-in Price (OG) MKR = $1956 (11/12/20 - 05/12/21)
Biannual MKR Amount Initial Annual Incentive Value (USD) / MKR/USD lock-in Price
Vesting Period 4 years
Cliff Vest 12 months
Vesting Schedule After cliff has expired, the Biannual MKR amount vests every 6 months and is distributed on Jan/July 1st.
Manual Repricing yes
Auto-Renewal yes
Estimated Max Total team MKR After 1 Year (15.5 FTE)* 1051.25
Estimated Max Total team MKR After 4 Years (15.5 FTE)* 4205

*Note that this estimation is based on all 15.5 FTEs joining the Oracle Core Unit on day 1. There are currently, only 6 FTEs (potentially 5) working for the Maker Foundation that are joining the Oracle Core Unit which significantly reduces the actual amount of MKR expended for Oracle Core Unit vesting. While we will try to begin hiring as soon as possible, we’re absolutely prioritizing to find the highest caliber candidates to fill vacant positions and will take our time accordingly. In the meantime the trailing 6 month average MKR price should rise which will reduce the amount of MKR awarded to new hires.

Manual repricing allows any contributor of the Oracle Core Unit to calculate a new MKR/USD lock-in price using the trailing 6 month average. This ensures that contributors who join during a bull market aren’t penalized relative to new contributors who join later during a bear market. In order to prevent abuse, manual repricing will reset the 12 month cliff vest period.

Related Documents

MIP39c2-SP12 Oracles Core Unit MIP

MIP41c4-SP13 Facilitator Onboarding, Oracles Core Unit Proposal

Oracle Core Unit Launch Pad Session


Approved!!! Including the 800K needed in case of emergency! Let’s goooooooooo


Love the detail Nik! I have a ton of question about Oracles but I’ll save those for the SP12 thread. Only have a few questions regarding the budget:

  1. What is a Feed Stipend?
  2. If you are unlikely to spend even half of the proposed budget within the year, why not set a shorter term budget (3-6 months) then request a new budget once you have better visibility into the back half of the year’s expenses?
  3. Is nearly 3x’ing the size of the team in a year feasible? Would your team have the capacity to onboard and train that many people while maintaining existing responsibilities and your sanity?
  1. Feeds are paid 1000 Dai per month for the service they provide. Given that we have 26 Feeds, this amounts to 26,000 Dai per month, and 312,000 Dai annually. This stipend is mean to cover the hosting costs as well as the engineering hours spent upgrading the Feed when the Oracle Domain Team (soon to be Oracle Core Unit) releases a new update. Frankly it’s rather low imo, but that’s a conversation for another day.

  2. I had considered and am still open to changing it if others feel strongly as well. My main concern is that I have no idea how many core units will join between now and 6 months from now, nor how much of a strain that will put on the finances of the protocol. It seemed safer to pitch the community and get you all invested in my vision, and assuage any doubts now when things are still rather quiet, instead of later when we’re in the thick of it and things are going at a million miles an hour.

  3. I agree it will be a challenge, but I think I’m up to it. I’ve been at Maker since the beginning and watched it grow from 7 people to 120+. I’ve seen what practices and processes can work exceptionally well in a small group of 5, and fail spectacularly in a group of 20, just as those that work for 20 are terrible for 50. Fortunately, while at the Foundation I’ve had the freedom to wear many hats simultaneously from engineering, to managing a team, to managing a product, business development, coordinating client integrations, managing invoices, writing blog posts, and even dabbling in social media. Not to mention, the incredibly high quality of the people around me I’ve been able to observe over the years.

I believe I’m ready :grin:


What is exactly is the service a feed provides? Is this just a feed for market/price data or? I have next to no knowledge of how oracles work as you may have surmised from my previous Chainlink questions in RC, ELI5 :slight_smile: .

Regarding number 2, that is a significant challenge for the DAO as a whole and one that @SebVentures and myself are working on providing visibility to. It’s very difficult to quantify what should be invested without having a budget or forecast for the protocol in place.

Each Feed is constantly sourcing prices of assets from different exchanges according to the Data Model that Maker Governance has ratified for each asset. The Feed then publishes a signed message containing the asset pair, price, and timestamp to a p2p network. So now you have this big pool of signed prices floating around on this network. A relayer comes along, grabs a bunch of these messages (enough to reach quorum) and pushes them to our Oracle smart contract. The smart contract verifies the signatures and then sets the median as the Oracle price. You can kind of think of the Feeds like a multi-sig of trusted actors that Maker Governance has appointed. The more Feeds you have, the more difficult it gets to launch an Oracle attack. Hence why instead of the $312,000 needed for 26 Feeds, we budgeted for $360,000 so we can add 4 more.


This is one of the very few cases where I’d happy to see money be spent fast! :stuck_out_tongue:

This is an amazing proposal. I feel this CU will bring enormous value to MakerDAO is the medium-long term (i.e., much more than what it brought already in the past).

Let’s gooo!
Screenshot 2021-05-13 at 11.00.15
Screenshot 2021-05-13 at 11.01.57


The actual proposal isn’t out yet though :stuck_out_tongue:


the budget proposal. I am waiting for the content proposal too! :slight_smile:

I am moving this MIP into Formal Submission. @charlesstlouis @Davidutro @blimpa


Question on this bit:

Note that the Maker Protocol retains full administrative control of the funds in the Oracle Core Unit Multisig .

This seems to have been accomplished by giving governance an approval to claw back these funds via the Gnosis Safe Allowance Module–see this transaction. However, the multisig signers can reset this allowance at any time with no delay. Thus if the intent is to preserve clawback rights for the DAO in case the team goes off the rails, I don’t think this solution works. The way it’s set up is still effective for clawback if e.g. enough signers lose their keys that the multisig is bricked, but it’s not effective against malicious activity. If this is intended I see no issues, but wanted to call it out just in case there was any confusion around this point.


As part of our broader effort to bring more transparency to the CU budget structure, we have documented the wallet setup of this CU and others;

Read more about it here: Introducing the CU Budget Transparency Map

1 Like

@NikKunkel how many people have you successfully onboarded into the Oracle team and when were their initial start dates?

1 Like