After Building 400+ Production AI Agents: 7 Skills Every AI Engineer Needs
MaryamMiradi · x · 2026-11-19
AI engineer Maryam Miradi distills lessons from building 400+ production AI agents into seven recurring skills:
- System design — agents are multi-component systems; structure data flow and isolate failures.
- Tool & contract design — vague tool descriptions make LLMs guess; use strict input/output types, valid examples, and validate inputs before execution. Treat tool schemas like legal contracts.
- Retrieval engineering — chunk size, document quality, and ranking make or break RAG; bad context beats even the best LLM. Learn chunking, hybrid search, and reranking.
- Reliability engineering — retries, timeouts, fallbacks, and circuit breakers; backend engineers solved this years ago.
- Security & safety — validate inputs and outputs, enforce least privilege on tools.
- Evaluation & observability — trace every tool call, test against known results, retest after changes; a great demo isn't a reliable product.
- Product thinking — show what the agent did, surface uncertainty, let users review critical actions.
Bottom line: skip chasing every new framework — tools change, systems stay.
More from coding & agent
- Eve agent tool cuts build time 8x to 6.9s by reusing compiler module graph — evilrabbit_ · 2026-11-19
- Dev shares NL-driven testing UI: 'the best interface is chat' — KlausCodes · 2026-11-19
- Devs flag GPT's defensive programming for hiding bugs; OpenAI asks for /feedback cases — pvncher · 2026-11-19
- Flutter ships A2UI 1.0: agents respond with real interactive widgets, not chat text — rseroter · 2026-11-19
- Skills vs MCP vs RAG vs Memory: the four context mechanisms AI agents actually need — MaryamMiradi · 2026-11-19
- Anthropic details agent memory: 'dreaming' lets Claude organize its experience — daniel_mac8 · 2026-11-19