Agentic Engineering: AI as Junior Dev via Context & RPI Loop

Treat coding agents as fast but judgment-lacking junior devs: master context engineering and research-plan-implement workflow to gain 30%+ time savings without quality loss.

Mental Model: AI Agents as Enthusiastic Junior Developers

Brendan O'Leary reframes coding agents not as autocomplete tools but as collaborators akin to junior engineers. Evolved from 2020s line-finishers to 2026 executors that break down tasks, edit files, run tests, and create PRs. This shift demands treating them as "energetic enthusiastic extremely well-read often confidently wrong junior developers"—fast, tireless, ego-free, with vast knowledge across languages/frameworks, but lacking business judgment or architectural context.

Arman, Flask creator, gained >30% daily time by directing handoffs: "we're no longer just using machines we're now working with them." O'Leary stresses articulating workflows—what to hand off vs. keep—to bridge the gap where 90% of engineers use AI but few maximize it. Blind acceptance yields "technically correct and contextually wrong" code; direction amplifies human thinking.

Quote: "think about your AI agent as an energetic enthusiastic extremely well-read often confidently wrong junior developer" (O'Leary's core mental model, explaining why agents excel at speed/breadth but fail on nuance, urging judgment as the human edge).

Context Engineering: The Art of Selective, Isolated Inputs

Context is the linchpin: expensive (tokens compound costs), degradable (quality drops >50% window fill), and poisonable (bad/outdated/mixed inputs corrupt outputs). MCP servers auto-load context, pushing into "dumb zone." Solutions: persist externally (scratchpads, agents.md), select relevant slices (file @mentions, disable unneeded MCPs), summarize/trim post-deep dives, isolate via new sessions or parallel agents.

O'Leary's intern anecdote illustrates: Wireframed iPad patient-history app in Balsamiq (Comic Sans, emoji placeholders) handed to interns yielded literal prototype. Fault: poor context curation, not juniors. Same for agents—"not giving the right context... what's important what's not."

Habits: One task/session, monitor context meter, restart with agent-written summary prompt if off-rails. Karpathy: "context engineering is a delicate art and science." Enables task separation, mirroring junior eng management.

Quote: "more context doesn't always mean better results... it can make the model actually dumber" (Highlights quality-cost tradeoffs, why selective isolation beats dumping everything).

Research-Plan-Implement Workflow: Leverage Human Thinking Upfront

Avoid "help me implement X" pitfalls—jumping to code assumes wrong, wastes time, breeds anti-AI sentiment. Instead, RPI loop:

  • Research (Ask Mode): Non-executable chat-only (Kilo's "ask mode" reads files optionally). Understand codebase, data flow, paradigms, edges. Brainstorm. Output: reviewable doc aligning human/AI understanding.
  • Plan: Explicit steps—files touched/created, verification tests, in/out scope. Output: step-by-step plan.md (common in repos). Use cheaper models here.
  • Implement: New session with plan only. Low context, frequent Git commits (O'Leary's GitLab bias: local Git as "first PR review"). Human review each change.

Human leverage max in research/plan; Dexory: "a bad line of research can potentially be hundreds of lines of bad code." "AI can't replace thinking it can only amplify the thinking you've done." Skips demo-style code-spew; see path.lo.ai for patterns.

PhaseGoalTools/OutputsHuman Role
ResearchUnderstand systemAsk mode → research docReview/align
PlanOutline changesPlan.md w/ steps/tests/scopeHigh-leverage thinking
ImplementExecuteCode mode + Git commitsApprove/iterate

Quote: "AI can't replace thinking it can only amplify the thinking you've done or the lack of thinking you haven't done" (Dexory via O'Leary; justifies RPI's upfront investment for reliable execution).

Agent Configuration: Modes, Rules, and Custom Playbooks

Tailor via modes (Kilo: ask/code/architect for role-focus), workspace rules (build/test commands, testing reqs), tunable autonomy (auto-approve reads/tests? Parallel agents? Worktrees?). Buckets:

  • agents.md: De facto standard—always-loaded README: conventions, commands, reqs.
  • skills.md: On-demand playbooks (e.g., changelogs, motion graphics)—reusable workflows.

Power tips (Kilo/VS Code): @mention files/commits, /commands (new task, condense context), select-code right-click. Tune as you learn; start conservative.

Iterate comfort: Begin low autonomy, expand. Git for safety nets pre-PR.

Quote: "a bad line of research can potentially be hundreds of lines of bad code" (Dexory; underscores why specialized modes/rules prevent implementation disasters).

Key Takeaways

  • Adopt junior dev mental model: Hand off grunt work, retain judgment/context.
  • Monitor context <50% fill; persist/select/summarize/isolate to cut costs/degradation.
  • RPI loop: Spend human time on research/plan for 30%+ gains; implement in fresh, low-context sessions.
  • One task/session; restart with agent summaries if derailed.
  • Mandate agents.md (rules/conventions); use skills.md for repeats.
  • Frequent local Git commits as agent "PR review."
  • Modes limit scope: Ask for research, code for execution.
  • Tune autonomy gradually; @mentions//commands accelerate.
  • Check path.lo.ai for workflows; avoid code-first prompts.
Video description
Coding agents are quickly moving from novelty to necessity, but most teams are still stuck between demos that feel magical and systems that break down in real-world engineering environments. In this session, Brendan O’Leary explores what it takes to make coding agents reliable collaborators rather than unpredictable copilots. Drawing from hands-on experience building and scaling AI coding agents, Brendan can unpack where agents succeed, where they fail, and how engineers can design workflows that balance speed with control. Attendees will learn how to think about agent autonomy, context management, and human-in-the-loop design so AI can meaningfully accelerate development without sacrificing code quality, security, or trust. This talk is for engineers ready to move past “vibe coding” and into production-grade agent-driven software development. Brendan O'Leary - Developer Relations Engineer, Kilo Code As conversations shift from AI demos to real engineering and coding agents begin moving into production environments, Brendan is passionate about helping teams understand not just what’s possible, but what’s practical. He’s especially energized by audiences who are grappling with the same questions he sees every day: how much autonomy to give agents, how to keep humans meaningfully in the loop, and how to move beyond “vibe coding” into reliable software development. Brendan is a builder and practitioner at Kilo Code, working hands-on with AI coding agents and the realities of deploying them in serious engineering contexts. He’s mastered the role of choreographer, successfully balancing the collaborative dance between human creativity and machine capability. His perspective of coding agents is rooted in lived experience, combining a deep technical understanding with a clear-eyed view of where agents succeed, where they fail, and why trust is the missing layer most tools overlook. Brendan brings a candid, engineer-first approach that resonates with technical audiences and leaves them with concrete ways to rethink how humans and coding agents collaborate in production systems. Socials: https://www.linkedin.com/in/olearycrew/ https://boleary.dev/ https://x.com/olearycrew https://gitlab.com/brendan/boleary-dot-dev https://kilo.ai/

Summarized by x-ai/grok-4.1-fast via openrouter

8137 input / 1863 output tokens in 12587ms

© 2026 Edge