A seamless way to pull in all communications from all the services I'm on:

• 20+ Slack Groups
• 10+ Telegram Groups and Conversations
• 30+ WhatsApp Converstions
• 20+ Signal Conversations
• 50+ SMS Converstations
• Email
• LinkedIn
• Keybase
• Discord
• SwapCard
• Patron
• Whatever else I'm missing...

Perhaps there's already something like this, but I'm hesitant because it seems overwhelming and likely that it means I have yet another service to check...

Automated Networking Calls for EAs:

- EAs with calendly sign up for weekly random networking 

- the system randomly pairs based on calendly availability match

- the system leverages calendly to book meetings 

- EAs have meetings and if they find them useful, tell other EAs to sign up!

Once we have this working for EAs (for free or a nominal cost), we could roll out a paid, profit-maximizing version. If you build it, I'm happy to shill it far and wide and pitch it to investors!

A forecasting twitter bot

A bot you can call which notes a forecast and will remind you when it's up and ask if it resolved positive or negative. It screenshots the tweet and stores it in a list of your forecasts. When the time is up it asks you if you were right or wrong and records this too.

If asked by someone else it will give your calibration. If people lie about their forecasts that will be very easy to find. 

Nuño Sempere and I both came up with this idea and he's currently applying for funding.   https://twitter.com/metaforecast/status/1459606291169652743?s=20

The two EA/Rationalist friend matchmaking apps should work together. 

https://www.reciprocity.io/ does facebook

https://tweetledee.xyz/ does twitter

Both are built by EAs. There should be a single app that lets one put in their facebook and twitter details and matches friends or anyone mutually present on twitter lists.

More frienships/relationships = good.

I would appreciate a code review of Metaforecast (front-end, back-end).

There's quite a few opportunities I see from looking around in EA. I am doing direct technical work for EA right now.

EA CoLabs

EA CoLabs itself can be framed as a technical problem. It's the problem of optimally matching different skillsets to different projects to maximise utility. You could definitely tackle it from a fun technical perspective (say, using the Hungarian Algorithm for matching, and using the Australian Skills Classification to describe skills). These however are just my ideas. I may be currently too busy with other things to properly investigate whether this is feasible.

Certificates of Impact and EA Economies

There are a lot of interesting ideas in creating economies around EA. For instance, having people sell products where all proceeds go to charity. I have a friend who wishes to look into using technologies such as blockchain to track coins that can be used to purchase good where the profits end up to Effective Charities.

Improving Infrastructure around Cost Effectiveness Analysis

This is what I'm currently doing as direct work. Cost Effectiveness Analysis are not really done to a very high quality in EA spaces as yet, and it would be nice to see more analysis so that we can make more informed decisions. I see this as a major tooling opportunity, as the state of the art in this area uses excel and a lot of time. This may include improving on technologies like Guesstimate, or building entirely new technologies.

Improving Infrastructure around epistemics and forecasting

The next area I would highlight is that there needs to be better evaluations of things like Cost Effectiveness Analysis, or really any other important calculations or predictions. These evaluations often aren't really done, and I would love to see proper evaluations say for parameters in GiveWell's Cost Effectiveness Analysis.

If you are interested in any of these. Please reply/get in contact.

Easy, Two-Way Opt-in Intros for Match-makers:
-a web service where I enter the e-mail addresses for 2 EAs I think should meet (to collaborate on similar ideas; it's not a dating app!)
-each EA receives a personalized email saying that I think they should meet the other and why
-recipients can accept, decline, ignore or opt-out completely
-only if both recipients accept in a reasonable time frame (say 1 week), does the system generate an automated email to connect them (and possibly auto-schedule if they both have calendly -- see above)

As the match-maker, I get feedback based on who is accepting/declining the requests. I know who in my network thinks my intros are actually worthwhile. I also save lots of time and avoid missed connections from email overload

Bounty system for stuff like this.

A website with findable bounties for EA goals. If you do the goal you get the bounty.

EA Tech Initiatives

https://docs.google.com/spreadsheets/d/1yzmg02j8PnvjlV_KX3Vf_u9TNdAS63N-9xy2txmTg-A/edit#gid=957019158

H/t Jaime Raldua Veuthey

Personally I'm looking for someone to help me build a simple plugin for the Obsidian note taking app.

The plugin should generate a list of links to notes that match criteria I specify.

Spec here. If you'd enjoy getting paid to make this for me, please send me a DM.

Filter the forum articles which get sent to me by email from the EA forum.

Just as the forum front page can be sorted by subtag, it would be great if newsletters could too. Ideally if RSS worked too.

Using the EA forum to aggregate and prioritise jobs from twitter/80k.

Twitter + 80k -> forum -> twitter + email/RSS

All 80k jobs and jobs tagged with @effective_jobs get scraped and posted to the forum (hidden). You can hear about specific jobs from the forum with an email or RSS.  The bot then posts all the ones that receive a certain amount of karma back to twitter.

I've written a much simpler Minimal Viable Product here: https://forum.effectivealtruism.org/posts/uxfWrFNH7jSSGhkkS/unofficial-pr-faq-posting-more-jobs-to-the-forum

A fake fund which is decided using this forum the scored by an actual fund 

Users have some impact standard and score projects. Any projects then get scored by one of the actual funds. How good were those suggested by the forum alone?

The crowd picks higher variance options when the greenlight plays (see below). If that's true for funding, the gains will far outweight the losses. Have we tested if this is the case?

Automating processes for community building.

Excerpt from this Google Doc https://docs.google.com/document/d/15ES8_Ah5hWjaiwprGfZpQMHx0AmfDGam6rAWWRGb6lY/edit:

• automating fellowship application solicitation and automating fellowship cohort matching process (just as MIT does dorm matching). automate airtable.
• writing emails automatically. pick a template (from a giant decision tree on a wiki), automatically send to a list of emails, fills in title, time, date.
• automatic discussion group setups
• automatically send followups with mail merge, etc. to keep in touch with members
• some sort of easily extendable platform for sending emails about events, meetings, projects
• organizing all of the EA hub materials for clubs into some place where people actually see them and can save time
• How can we make it possible to set up an event in 5 minutes total?

A scheduling service that works across multiple timezones.

Specifically for organising seminar series with 100s of participants and 10s of facilitators. We need to:

• Take in facilitator & participants' availabilities, submitted by them in their timezone
• See, for each participant, which facilitators are available at this time. (So that we can match appropriate facilitators to the knowledge status of each participant, constrained on availability.)
• Preferably some adjustment for upcoming daylight savings, but that's a stretch.

Current solutions: when2meet doesn't sync across timezones for Mon-Sun availabilty (only specific dates). Lettucemeet has weird overflow errors around midnight, and does not work in India (+4.5 hours, I believe).

This problem is known to some people, and I don't know the status of solving it at time of writing the comment (I expect the work has not been started though). Do get in touch if you're interested in this problem before going ahead with anything, all the same!