Some more things about Django I've been enjoying

The author’s current project uses Django to serve a multi‑page site that relies on an SQL backend and minimal JavaScript. By defining custom QuerySet methods—such as approved(), future(), and with_tags()—they chain filters in view code, turning raw SQL into readable, reusable Python. Template‑side conveniences like urlize, linebreaksbr, date formatting, and the json_script filter streamline HTML generation, while the rarely‑discussed querystring filter lets the author tweak URL parameters without manual string handling. Automatic migrations have already produced 19 schema changes, underscoring Django’s “edit‑model‑run‑migrate” workflow that accelerates iterative development.

The author’s experience reflects a broader swing among developers who grew accustomed to static site generators, Vue SPA front‑ends, or micro‑frameworks like Flask and Go’s stdlib, and now seek the productivity of a mature, batteries‑included framework. Django’s emphasis on convention over configuration contrasts with the author’s earlier “backend‑light” approaches, offering a single language stack that reduces context switching. Their aversion to class‑based view inheritance mirrors a growing discourse favoring function‑based views for clarity, especially among teams wary of deep inheritance hierarchies. Meanwhile, the performance snapshot—2‑3 requests per second on a cheap VM, with occasional bursts from LLM scrapers—highlights the latency gap between Python‑based stacks and compiled alternatives such as Go, prompting questions about scaling, caching, and templating choices.

Looking ahead, the author must decide whether to invest in profiling tools like py‑spy to pinpoint bottlenecks or to adopt higher‑level mitigations: enabling Redis or Memcached layers, moving static assets to a CDN, or swapping Django’s default templating engine for Jinja2, which the official docs claim is faster. The trade‑off between template {% block %} versus {% include %} optimizations may be marginal, but systematic caching (per‑view, template fragment, or low‑level query caching) could lift throughput enough to handle sporadic traffic spikes without over‑provisioning the VM.