https://www.dreamclerk.com/blog field notes from building dreamclerk. the data, the rejects, the rubric changes. nothing polished after the fact. en-IN 2026-07-28T09:00:00.000Z https://www.dreamclerk.com/blog/why-we-built-dreamclerk https://www.dreamclerk.com/blog/why-we-built-dreamclerk 2026-04-12T09:00:00.000Z dreamclerk team we ran a 90-second phone screen on 1,200 internship applicants. 14% got a callback. here is what we changed, and what we got wrong. founder notes internship https://www.dreamclerk.com/blog/inside-our-bias-audit https://www.dreamclerk.com/blog/inside-our-bias-audit 2026-05-04T09:00:00.000Z dreamclerk team every quarter we run a bias audit on the dreamclerk interview. here is the rubric, the per-group pass-rates, and the three rubric changes we made after cohort 1. bias audit hiring https://www.dreamclerk.com/blog/coding-interview-with-no-experience https://www.dreamclerk.com/blog/coding-interview-with-no-experience 2026-05-04T09:00:00.000Z dreamclerk team no internships. no github. no leetcode streak. here is the 4-step protocol we built to prep dreamclerk applicants in 6 weeks — and what the data says about it. interview career https://www.dreamclerk.com/blog/in-browser-ide-explained https://www.dreamclerk.com/blog/in-browser-ide-explained 2026-06-10T09:00:00.000Z dreamclerk team monaco on top of a webcontainer. 200ms cold start. 12mb of wasm. here is the architecture, the security model, and the 3 things it cannot do. engineering explainer https://www.dreamclerk.com/blog/shipping-code-vs-knowing-code https://www.dreamclerk.com/blog/shipping-code-vs-knowing-code 2026-06-10T09:00:00.000Z dreamclerk team the 14 terms every dreamclerk applicant should be able to use in a sentence, with a worked example for each. bookmark this. re-read it before the interview. glossary engineering culture https://www.dreamclerk.com/blog/fresher-unemployment-india-2026-the-numbers-and-the-fix https://www.dreamclerk.com/blog/fresher-unemployment-india-2026-the-numbers-and-the-fix 2026-06-16T09:00:00.000Z dreamclerk team 73% of indian engineering graduates are unemployed a year after college. we break down the 2026 numbers, the four reasons companies reject freshers, and the one fix that is actually moving the needle. fresher hiring unemployment https://www.dreamclerk.com/blog/the-2-year-experience-trap https://www.dreamclerk.com/blog/the-2-year-experience-trap 2026-06-19T09:00:00.000Z dreamclerk team 85% of indian tech JDs ask for 2+ years experience. only 12% of applicants have it. the gap is structural, not preference. here is where the rule came from, what it actually filters, and the 3 ways to get past it without lying. experience hiring https://www.dreamclerk.com/blog/how-to-get-hired-as-a-fresher-with-no-internship-and-no-network https://www.dreamclerk.com/blog/how-to-get-hired-as-a-fresher-with-no-internship-and-no-network 2026-06-22T09:00:00.000Z dreamclerk team no internship. no github. no alumni network. no tier-1 college. here is the exact 6-week protocol that 187 dreamclerk applicants used to go from 14% interview rate to 31%, with the calendar, the artifact list, and the rejection log included. playbook fresher hiring https://www.dreamclerk.com/blog/why-2-years-experience-required-is-a-tax https://www.dreamclerk.com/blog/why-2-years-experience-required-is-a-tax 2026-06-25T09:00:00.000Z dreamclerk team a hiring manager writes. the rule is a fossil. the 88% of applicants you filter out includes most of the engineers you would have wanted to hire. here is the data, the math, and the 4-step replacement. hiring manager tax https://www.dreamclerk.com/blog/the-resume-is-dead-three-signals https://www.dreamclerk.com/blog/the-resume-is-dead-three-signals 2026-06-28T09:00:00.000Z dreamclerk team the resume predicts 6-month retention at r=0.12. three other signals predict it at r=0.40, r=0.38, and r=0.34. here is what they are, why they work, and how to build all three in 90 days without a tier-1 college or a brand-name internship. signals hiring https://www.dreamclerk.com/blog/what-a-good-postmortem-looks-like https://www.dreamclerk.com/blog/what-a-good-postmortem-looks-like 2026-07-15T09:00:00.000Z dreamclerk team a good postmortem takes 30 minutes to write. it has 5 sections. it is blameless. it is dated. it has follow-ups that are assigned and dated. here is the worked example, the template, and the 3 sentences you should never write in one. postmortem engineering culture https://www.dreamclerk.com/blog/india-it-services-the-fresher-pipeline-is-broken-and-it-is-not-the-colleges-fault https://www.dreamclerk.com/blog/india-it-services-the-fresher-pipeline-is-broken-and-it-is-not-the-colleges-fault 2026-07-13T09:00:00.000Z dreamclerk team tcs, infosys, wipro, and hcl hired 1.4 L freshers in fy24, then 87,000 in fy25. the 84% drop is not a market correction. it is a structural break. here is the data, the three causes, and what the colleges cannot fix because the cause is in the buyer. india it services https://www.dreamclerk.com/blog/building-the-in-browser-ide-a-postmortem https://www.dreamclerk.com/blog/building-the-in-browser-ide-a-postmortem 2026-07-10T09:00:00.000Z dreamclerk team we shipped a regression in june 2026 that made the in-browser ide cold-start 4.2s instead of 200ms. here is the postmortem: timeline, contributing factors, root cause, what went well, follow-ups. dated, blameless, public. postmortem engineering https://www.dreamclerk.com/blog/the-cost-of-a-bad-junior-hire-is-not-the-onboarding https://www.dreamclerk.com/blog/the-cost-of-a-bad-junior-hire-is-not-the-onboarding 2026-07-08T09:00:00.000Z dreamclerk team we ran the numbers on 47 bad junior hires across 12 indian mid-size startups in 2024–25. the onboarding is 18 LPA. the 6 months after is 38 LPA. here is the breakdown, the math, and the 4 questions that would have caught 41 of the 47. hiring roi https://www.dreamclerk.com/blog/why-we-stopped-using-take-home-projects https://www.dreamclerk.com/blog/why-we-stopped-using-take-home-projects 2026-07-06T09:00:00.000Z dreamclerk team 6-hour take-home projects are the most expensive, least predictive signal in 2026 hiring. the data: r=-0.04 with on-the-job performance, r=0.31 with AI-assistance. here is what we replaced them with, and the 90-second alternative that works. interview take-home https://www.dreamclerk.com/blog/the-intern-cheat-sheet-for-the-pushback-round https://www.dreamclerk.com/blog/the-intern-cheat-sheet-for-the-pushback-round 2026-07-03T09:00:00.000Z dreamclerk team the pushback round is the part of the dreamclerk interview with the highest correlation to cohort retention. here is how to prep, how to run one with a peer, and the 4 sentences that will save you when you are losing. interview pushback https://www.dreamclerk.com/blog/the-dual-path-review-engine https://www.dreamclerk.com/blog/the-dual-path-review-engine 2026-07-01T09:00:00.000Z dreamclerk team we wrap our deterministic regex engine with a claude haiku call. the haiku is constrained to the same rubric. if the call fails, the regex returns the verdict. here is the architecture, the contract, the quota, and why both paths are kept. ai review https://www.dreamclerk.com/blog/the-2026-pr-review-is-async-and-warm https://www.dreamclerk.com/blog/the-2026-pr-review-is-async-and-warm 2026-06-29T09:00:00.000Z dreamclerk team the 24h pr was a fossil from a 2014 office culture. the 2026 pr is async-first, written, and has a warm handoff. here is the playbook, the data, and the 4 things you have to stop doing in your engineering org. pr review engineering culture https://www.dreamclerk.com/blog/the-first-90-days-at-your-first-tech-job https://www.dreamclerk.com/blog/the-first-90-days-at-your-first-tech-job 2026-07-22T09:00:00.000Z dreamclerk team the first 90 days decide whether you pass probation or get pip'd. here is the one thing to do in week 1, week 4, and week 12 — the protocol that took us 18 months to learn and 12 minutes to read. fresher first job https://www.dreamclerk.com/blog/6-pip-signals-and-how-to-flip-4-in-30-days https://www.dreamclerk.com/blog/6-pip-signals-and-how-to-flip-4-in-30-days 2026-07-25T09:00:00.000Z dreamclerk team pip decisions are not made in pip meetings. they are made 6–8 weeks earlier, from 6 observable signals. here is the list, which 4 are flip-able in 30 days, which 2 are structural, and the one sentence that gets you a re-rating. fresher pip https://www.dreamclerk.com/blog/off-campus-hiring-2026-the-7-channels-that-still-work https://www.dreamclerk.com/blog/off-campus-hiring-2026-the-7-channels-that-still-work 2026-07-28T09:00:00.000Z dreamclerk team 78% of indian tech hiring is off-campus in 2026. of the 11 channels freshers use, 7 still work and 4 are dead time. here is the data, the 7 ranked by signal-to-noise, and the one we have never seen anyone fail. fresher off-campus