I build the working prototype, not just the spec.

@@ −2026.02 +2026.10 @@ asifmk.com

1. 5Added: A changelog of my career, built under two Claude skills: Impeccable for design and my own writing standard for copy.built with AI
2. 6Added: Production from one Cursor prompt: a private GitHub repo, Cloudflare Pages, a deploy on every push, and a custom domain.built with AI
3. 7Changed: Version two after reading the live site as a hiring manager would: plain-word labels on every line, LinkedIn detail, and deeper release notes.
4. 8Removed: Left out an AI chat widget. Shipped work tells a hiring manager more than a chatbot does.

@@ −2025.12 +2026.02 @@ ReturnKoto?

1. 9Added: 169 public pages: a calculator, 38 fund pages, 90 head-to-head comparisons, weekly data reports, guides, and a FAQ.built with AI
2. 10Added: A data pipeline with its own parser for each of 10 fund managers' websites. It validates every price before the site rebuilds itself.built with AI
3. 11Added: A publish-or-withhold rule: a figure is either right or not shown. Drift detection stops a bad sync before it reaches the page.
4. 12Fixed: A fund manager reported our numbers as wrong. The audit found month-end dividends falling outside the look-back window. After the fix, the manager's own published returns reproduce to within 0.01 percentage points.
5. 13Added: A Claude skill that holds every post and page to one writing standard: no unverified figures, no vague authority, no filler.built with AI
6. 14Withheld: Fund B, 5-year return: 00.0%figure not shown, withheld: two sources disagree. Illustration of the rule, not live data.

@@ −2024.11 +2025.12 @@ InsideMaps

1. 15Added: Launch scope, launch criteria, and the one metric that decides whether a pilot continues, across 4 regions.
2. 16Added: Working prototypes built with Claude from customer problems, each with a short written recommendation. Engineering owns how it gets built.built with AI
3. 17Changed: Product roadmap tied to commercial targets with the executive team, including where to expand next.
4. 18Removed: A feature concept that fails as a prototype stops there, before it costs engineering time.

@@ −2021.07 +2024.11 @@ Deloitte

1. 19Added: Two AI workflow products taken from prototype to production. I owned discovery, priority, and rollout.built with AI
2. 20Removed: Cut manual processing by 20%.
3. 21Added: AI proofs of concept with business cases and ROI frameworks, so executives could decide where to invest. Model and architecture trade-offs settled with engineering.built with AI
4. 22Added: The problem statement, the eval bar, and the kill criteria, written before anything was built.
5. 23Changed: AI readiness assessments and build-versus-buy calls for clients deciding what to adopt and what to fund.
6. 24Added: Strategy workshops with international clients and regulatory stakeholders, so the technical roadmap met compliance and go-to-market needs.

@@ −2020.08 +2021.07 @@ SELISE Group

1. 25Fixed: Third-party API integrations failing. Success rate up 45% after I wrote the PRDs, API contracts, data mappings, and acceptance criteria partners shipped against.
2. 26Removed: Cut partner support tickets by 60% by automating the manual handoffs where integrations broke.
3. 27Removed: Cut time-to-market by 25% across four enterprise platforms, including Subway, ViCAFE, and Berger Paints, with one shared backlog across engineering, QA, and design.
4. 28Added: 3 CMS platform launches on WordPress and Laravel, from requirements to production.
5. 29Added: API documentation and developer onboarding material for partner engineering teams.
6. 30Changed: Recurring regional requests merged into one repeatable platform backlog.

@@ −2019.11 +2020.08 @@ Sheba Platform

1. 31Added: Payments for 150K+ micro-entrepreneurs on iOS and Android, shipped to 90% rollout, in a market of 165M+ users.
2. 32Added: The sManager Online Store: an onboarding flow that let 500K+ small businesses set up a working online storefront in under 60 seconds.
3. 33Changed: Sheba.xyz revamped from evidence: Amplitude funnels, A/B tests, and 30+ in-depth user interviews.
4. 34Added: 100+ customer interviews turned into personas, journey maps, and information architecture.
5. 35Changed: A change shipped only after it moved activation or payment conversion.

@@ −2018.12 +2019.11 @@ SWAP

1. 36Added: The product, from persona mapping and discovery workshops to App Store launch, with a 23-person R&D team.
2. 37Added: Engagement and product-market-fit evidence that helped the company raise a $1.25M seed round in October 2021.

@@ −2016.07 +2018.12 @@ GameOf11

1. 38Added: A first release within six months, then the app from zero to 100K+ users on iOS, Android, and web.
2. 39Added: User lifetime value up 28% after reworking onboarding flows and game mechanics.
3. 40Added: Coin packs from 20 BDT to 500 BDT and up, so a new player could start small and a regular could buy in bulk.
4. 41Changed: The referral reward pays out only after the new user enters a contest, so the incentive buys players, not empty signups.
5. 42Removed: Dropped reward mechanics that did not move Day-7 or Day-30 retention or LTV, found through A/B tests in Amplitude, Mixpanel, and SQL.
6. 43Added: A funnel scorecard of DAU/MAU, conversion, retention, and LTV that set the order of the roadmap.
7. 44Changed: The media budget, mostly Meta ads plus local football tournament sponsorships, held to one bar: did the user play.

@@ +2016.07 @@ Flow Digital

1. 45Added: Multi-year roadmaps for a portfolio of B2B SaaS and consumer web and mobile products.
2. 46Added: Flutter mockups for rapid prototyping, handed from design to senior engineering as working screens.
3. 47Added: Competitive teardowns, user research, and market analysis that decided what got built next.
4. 48Added: Interaction concepts and prototypes for client pitches, and heuristic reviews before each release.