TLDR is the free daily email that does the reading for you. Each issue summarizes the most interesting stories in startups, tech, and programming, curated by ex-Google and Anthropic engineers, so you get the signal without the noise.

It's one email instead of 12 open tabs. A 5-minute read with your morning coffee, and you walk in already caught up.

Tech is just the start. We also cover AI, marketing, dev, and more, so you can follow the topics that matter to your work.

Free, daily, and read by 7M+ subscribers. Subscribe and get your mornings back.

Subscribe for Free

You polish the resume. You tailor the cover letter. You get the email back, and instead of an interview it is a link to a skills assessment, or a take home project, or a timed test you have 48 hours to complete. For a lot of jobseekers right now, that link is where the real decision gets made, and most people treat it like a formality on the way to the "real" interview. That is backwards.

Skills first hiring stopped being a buzzword. By 2026, surveys put the share of employers using skills based hiring around 85 percent, with roughly 81 percent running specific skills assessments, up from about 56 percent in 2022. A growing chunk of companies now put a test in front of you before a human ever reads your resume. The thing you spent all weekend perfecting is no longer the gate. The assessment is.

And here is the part almost nobody has caught up to: the assessment itself is mid transformation, because AI broke the old version of it. The take home you could quietly run through ChatGPT is on its way out, and what is replacing it rewards a completely different kind of preparation. Get this stage right and you walk past people with stronger resumes. Get it wrong and the strongest resume in the pile does not save you. Here is how it actually works now.

The mental model most people carry is resume gets the interview, interview gets the job, and the skills test is a box to check somewhere in between. In 2026 that order is collapsing. Roughly four in ten organizations are shifting toward skills assessments over traditional interviews as the thing that actually decides, and many screen on the test first and read your background second, if at all.

What that means in practice: the hour you would have spent on a fifth resume tweak is far better spent preparing for the assessment stage. Know the common formats for your field (a work sample, a timed problem set, a case, a portfolio review), and walk in expecting to be measured on what you can do, not what you claim. The candidates getting through are the ones who realized the test is the interview, and prepared like it.

The async take home, the one you complete alone on your own time, is quietly being retired, and the reason is simple. Any candidate with basic literacy can now run a take home through an AI tool and submit polished work in minutes, so the test stopped measuring what it was built to measure. About 71 percent of business leaders say assessing technical skill has gotten harder because of exactly this.

So companies are moving to live evaluation. More first round filters now end with a short session where someone shares your screen, asks you to solve something in real time, or walks you through your own submission and asks why. Why this approach, why not that one, what would break at scale. An AI assistant whispering answers cannot survive a real time, multi turn "explain your reasoning" the way it can quietly write a take home. Prepare for that. The skill being tested is shifting from can you produce the answer to can you defend it out loud.

This is the trap that is sinking people right now. An analysis reported through SHRM, drawn from roughly 20,000 interviews between mid 2025 and early 2026, found signs of cheating in around 38 percent of candidates, rising to nearly half in technical roles, and a majority of hiring managers now say they suspect AI use during live assessments. The single biggest tell is the performance gap: you ace the written test, then fumble the moment someone asks a follow up about your own answer.

The fix is not to avoid AI. The frame has shifted from "did you use AI" to "can you actually do the job," and for many roles using AI well is the job. Use it to prepare, to learn the concept, to practice. But never submit something you cannot explain line by line, because the explanation is now the actual test. If you used AI to get to an answer, make sure you understand the answer well enough to defend every choice in it.

AI changed what senior engineering means. Gauntlet is built for engineers who want to lead that change — not watch it. The most effective way to become AI-first. No tuition. No tutorials. Just the work. Apply now.

Cohort is filling up now

_{Must be a US citizen to qualify.}

Not every take home is fair, and a tight market makes it easy to exploit you. A legitimate assessment is bounded and hypothetical: a contained problem, a clear time expectation, a stated rubric. The version to be wary of is the open ended one that hands you a real, current business problem with their real data and no limit, the kind of thing that looks suspiciously like the work they would otherwise pay someone to do.

Protect your time. Before you start, ask two questions: roughly how long should this take, and how will it be evaluated. A real process has answers. Then time box it to what they say and stop there, even if you could keep polishing. If an assignment is unbounded, asks for a complete solution to a live problem, or expects many unpaid hours, it is fair to propose a smaller scoped version or a paid trial instead. Wanting the role does not require you to work for free.

One more thing worth knowing: automated proctoring tools that watch your webcam or lock your browser produce false flags, and they disproportionately misfire on candidates with darker skin, disabilities, or a noisy home setup. If you are ever flagged, you are within your rights to ask for a human to review it rather than accepting an automated rejection.

The strongest position is making the assessment a confirmation rather than a hurdle. Skills first recruiters increasingly source straight from demonstrated work, so a public, verifiable body of what you can do is leverage. A GitHub with real projects, a portfolio of case studies, writing that shows how you think, a documented thing you built and shipped. When a recruiter can already see your work, you often skip steps, and when you do hit an assessment, you walk in having already proven the thing it is trying to measure.

This is the same shift showing up everywhere in hiring: proof beats claims. The candidate who can point to real work, and explain it on the spot, is the one the 2026 process is actually built to reward.

The window to prepare is narrower than you think. This briefing covers the verified signals that typically appear before a major IPO filing and what retail investors can legally access before a company goes public.

Get the roadmap