Manara.

The pipeline nobody saw coming.

It is a Tuesday morning in Berkeley, and someone in Gaza is being mocked by a Google engineer. Politely. On Zoom. About a binary tree question. This is, by any reasonable measure, a small miracle - and at Manara, it is also Tuesday.

Manara is an edtech company that trains software engineers, AI specialists and cloud architects across the Middle East and North Africa, and routes the best ones into jobs at companies most of them used to read about on Hacker News. It is funded by Stripe. It was incubated by Y Combinator. Its angel list reads like a LinkedIn screenshot somebody would post for clout: Reid Hoffman, Paul Graham, Eric Ries. The product is partly a bootcamp, partly a placement service, partly an Arabic-English AWS academy, and partly a very stubborn argument with the global tech industry.

A talent gap that wasn't.

For years, recruiters at global tech firms told each other a comforting story: there isn't enough engineering talent in the Middle East. The story was wrong. It is the kind of wrong that survives because nobody profits from disproving it. Computer science departments in Cairo, Amman, Ramallah and Beirut have been graduating women and men in respectable numbers for two decades. The graduates exist. The jobs exist. Connecting the two turns out to be a logistics problem, dressed up as a talent problem, dressed up as a geopolitics problem.

This is the central tension that runs through everything Manara does. Not "can we teach people to code?" - they already can. Not "can MENA produce engineers?" - it already does. The question is narrower and far more annoying: can a candidate from Khan Younis pass a phone screen at Meta? Can a self-taught backend engineer in Casablanca prepare for a Stripe loop in a second language, in a time zone eight hours off, without a single friend who has done it before?

Two women, one thesis.

Iliana Montauk and Laila Abudahi met around the same observation. Iliana had spent years running Gaza Sky Geeks, one of the earliest tech accelerators in the Palestinian territories - which is to say, she had a Rolodex of brilliant engineers who could not get past the third round of a FAANG interview for reasons that had nothing to do with the third round. Laila grew up in Gaza, became a software engineer at NVIDIA, and lived the gauntlet herself: Egypt, then Washington, then a real engineer's salary. She knew exactly which step on the ladder snapped under most people's weight.

Their bet, simply put, was that the missing piece was not curriculum - the internet has plenty of that - but apprenticeship. A senior engineer at Stripe spending forty-five minutes telling you the truth about a coding question. A peer in Amman doing mock interviews with you at 6 a.m. A coach who has read your resume the way a hiring manager will read it. Manara was built to be that apprenticeship at scale. It is, in retrospect, an obvious idea. Most great ideas are.

Less bootcamp, more bridge.

From the outside, Manara looks like a learning platform. Cohort-based engineering programs. Interview prep tracks. AWS Solutions Architect prep in Arabic. AI Practitioner. ML Specialty. Enterprise upskilling for governments that have decided their next major export will be software engineers. The catalog is conventional. The underneath is not.

What Manara actually sells is sequencing. The right mentor at the right week. The right peer at the right time zone. A resume reviewed by a person who has read four hundred resumes and remembers which words got someone hired at Datadog. The platform tracks where learners are in their job journey - applying, interviewing, negotiating - and matches them to the human who can move them one rung up. It is unglamorous, expensive work, which is precisely why nobody else does it well.

The numbers, in order of usefulness.

It is fashionable, in this corner of edtech, to put a hopeful number on a slide and hope nobody asks for a denominator. Manara is annoyingly forthcoming. The chart below is theirs, not ours. The denominators are real.

The customers list reads predictably for a company in this business: Google, Meta, Amazon and a long tail of mid-sized tech companies hiring Manara graduates. The partner list is the one to watch. AWS does not write a regional training partnership lightly, and 2,600 certifications is not a press release - it is a payroll.

A million lives, give or take.

Manara's stated goal is to impact one million lives in MENA. This is the kind of number that ought to sound like marketing and instead, when you spend ten minutes looking at the company's distribution, sounds a little conservative. Their content reaches engineers in Palestine, Jordan, Egypt, Morocco, Algeria, the UAE, Saudi Arabia, Lebanon, Tunisia, Iraq. Their certifications are denominated in dollars that flow back into local economies. Their alumni come back as mentors. The flywheel is annoyingly real.

The deeper mission is harder to put on a slide. It is about reshaping who the global tech industry thinks of when it thinks of "engineer." For a long time, that picture had a zip code. Manara is methodically updating the mental model. One mock interview at a time.

The next billion engineers.

The story of the next decade in software is not San Francisco's. It is a story about who gets to write the code that runs the AI that runs the world. If those engineers all come from the same three zip codes, the resulting software will be confidently wrong about a great many things. If they come from forty countries and forty languages, it will be less wrong, more often. This is not a moral claim. It is an engineering claim.

Manara is one of a small number of companies treating this as a serious technical problem rather than a charity opportunity. The cohort-based design, the AWS partnership, the mentor matching, the data-driven retention work - none of it is sentimental. It is logistics. It is the dull, expensive, unglamorous logistics of turning latent talent into deployed talent, at scale.

Back in Berkeley, the Tuesday Zoom ends. The Google engineer signs off. The candidate in Gaza closes her laptop, opens a notebook, and writes out the question again - this time, the way she should have answered it. In four months she will sit a real loop. In six, she will have an offer. In two years, she will be one of the mentors. The flywheel keeps spinning. The lighthouse keeps lighting.

The talent gap was never about talent. Manara figured that out first.