17 posts · notes, frameworks, guides, essays
What it actually looks like to wire Azure SQL, Azure ML, Cosmos DB, Azure Functions, and a Static Web App into one working system. Including the part where 82% accuracy means the model does nothing.
Dhaga & Co. had 44% of returns landing in a free-text field their category head could not scale through manually. Three agentic AI patterns turned that into actionable root cause intelligence for $2.50 a week.
A payroll SaaS company asked for a chatbot to cut headcount by 30%. The data told a different story. A walk through what problem discovery actually looks like in the field.
The prototype-to-production gap is not a model problem — it is a systems problem. Here are the four layers where AI deployments break after the demo looks great, and what to do about each one.
FDE interviews are deliberately open-ended. Most engineers freeze because they are waiting for a spec that never comes. Here is the framework that removes the freeze — a repeatable system for reasoning through ambiguous problems out loud.
The bottleneck in AI has moved from building to deploying. The engineers who win in this era are not the ones who write the best code — they are the ones who close the gap between a working model and a working system in a customer's hands.
Most AI personalization is obvious. The difference is in signal input, prompt design, and tone calibration. Here is the Clay + Claude workflow that produces personalization that reads like a human wrote it.
Claygent is the most powerful column in Clay and the easiest one to misuse. Here is how GTM engineers architect AI research workflows — model selection, prompt design, cost management, and validation queues.
80% of cold email success is infrastructure. SPF, DKIM, DMARC, domain age, warmup schedules, inbox rotation. Here is how GTM engineers build sending infrastructure that lands.
Most ICPs describe a market. A real ICP filters one. Here is the three-layer framework GTM engineers use — firmographic, technographic, persona — and why exclusion logic is the part everyone skips.
Signals are company-level events. Intent is person-level engagement. Triggers are automated workflows. Most outbound teams treat all three as the same thing — which is why their reply rates are 0.3%.
TAM is a ceiling. SAM is your reachable slice. SOM requires qualitative judgment — not just a smaller filter. Here is how GTM engineers use market sizing as a targeting architecture, not a slide deck number.
Most outbound teams think of their stack as a list of tools. Here is the mental model that separates engineers who build systems from operators who buy subscriptions.
No single data provider has complete coverage. Waterfall enrichment — layering Apollo, Hunter, Datagma, Dropcontact in sequence with conditional logic — is how GTM engineers fill the gaps without burning credits.
GTM engineering job listings grew 25x in 18 months. Here is what the role actually is, what it pays, and why the window to get in early is closing faster than you think.
Everyone rewrites their sequences. Almost nobody fixes their enrichment. That's the real gap.
FDE isn't a softer version of SWE. It's a completely different discipline — and the interview proves it.