AI Automation &AI Prompting: Unlocking Opportunities of the AI Economy

TechCrushHQ avatar
TechCrush.pro
@TechCrushHQJul 8, 2026

The Spaces convened a mentorship session on AI Prompt Engineering and AI Automation for Cohort 8 applicants, hosted by Abisola with speakers Michael Ajiboye (AI/ML and prompt engineer), Colin Eromosele Okunowo (prompt engineer, creative technologist), Chidera Okonkwo (AI automation specialist), and Favor (AI OS/workflows builder). Colin framed prompt engineering as the craft of precisely communicating goals to AI across design, content, automation, and operations, emphasizing consistency and mastery over dabbling. Michael traced a path from Python/data science to LLMs, highlighting refined prompting, evaluation, and business-focused results; he advised being grounded in clear objectives, remaining curious about tool capabilities, and acting as a “director” of AI tools. Favor detailed AI automation as converting manual processes into rule-based, AI-assisted workflows that understand requests, make decisions, and act (e.g., lead follow-up, CRM updates), and shared candid guidance on overcoming information overload, impostor syndrome, and the importance of positioning and business value. Chidera spotlighted building in public, landing clients, and how automation, CRM, agents, and no-code tools intersect, while warning that embedded agents raise the bar—urging learners to become broad “AI people” who solve problems end-to-end. A Q&A covered entry paths, tools, industries (education, media, customer support), human-in-the-loop limits, and practical career steps.

Mentorship Series: AI Prompt Engineering and AI Automation — Webinar Summary

Context and Objectives

  • Host: Abisola led a mentorship session for the Cohort 8 bootcamp to demystify two in-demand tracks: AI Prompt Engineering and AI Automation.
  • Core aims: clarify what each discipline entails in real practice, the skills needed beyond tool usage, whether they are sustainable career paths, how to spot and create opportunities, and how to position oneself for roles and projects.

Speakers and Roles

  • Michael Ajiboye — AI/ML Engineer and Prompt Engineer; background in data science, ML, LLM evaluation, and research lab work with frontier models.
  • Colin Eromosele Okunowo — AI Prompt Engineer and Creative Technologist; design/animation/content creation background integrating AI into creative workflows.
  • Chidera Okonkwo — AI Automation Specialist; transitioned from CS background and light UI/UX into no-code/low-code AI automation and voice agents.
  • Favor — AI Automation Developer; progressed from Virtual Assistant and Operations Manager to designing AI operating systems and workflows for businesses.

AI Prompt Engineering

What it is (beyond “typing prompts”)

  • At its core: structured communication with AI systems to reliably achieve a defined goal.
  • Practical scope:
    • Creative production (ideation, scripts, storyboards, graphic concepts) and content ops (summaries, proposals, executive briefs).
    • Product/design: synthesizing requirements, generating variations, speeding exploration.
    • Automation contexts: crafting system and task prompts inside agentic or automated flows.
    • Research/learning: extracting, structuring, and simplifying information.

Why it matters in the market

  • Businesses prioritize speed and consistency. Effective prompting bridges the gap between objectives and AI output quality, improving productivity and outcomes.

Skill vs. career path

  • Colin: It is a must-have skill across domains; can also be a sustainable career if paired with clear demand and implementation niches.
  • Michael: It becomes durable when tied to concrete value—evaluation, reduction of hallucinations, and domain-specific prompting that produces business outcomes.

Common beginner pitfalls

  • Colin:
    • Inconsistency: irregular practice prevents mastery, hence poor results and weak market demand.
    • Dilution: trying everything at once; spreading too thin across tools and subdomains.

Capabilities to develop (beyond tool familiarity)

  • Michael:
    • Grounding: be crystal-clear about the problem you’re solving and the audience/context.
    • Curiosity: continuously probe model capabilities, compare tools, and iterate prompts.
    • Director mindset: treat tools like “actors”—set roles, objectives, constraints, tone, and success criteria.
    • Evaluation literacy: refine prompts to reduce hallucination, compare model versions, and select tools that best fit the task.

Opportunity landscape

  • Michael and Colin highlighted opportunities across:
    • Content and marketing: ideation, scripts, calendars, briefs, brand voice consistency.
    • Design and creativity: concepting, style exploration, text-to-image/video assist.
    • Education: lesson plans, quizzes, simplification of complex topics.
    • Customer support: prompt logic for chatbots/agents; crafting behavior and responses.
    • Business operations: proposals, reporting, SOPs, documentation.
    • Training and mentorship: teaching teams/business owners how to get value from AI tools.

Where it’s headed (near term)

  • Continued diffusion into business workflows, automation stacks, and domain-specific tasks.
  • Prompt engineers who combine communication craft with evaluation and domain expertise will remain competitive.

AI Automation

What it is

  • Favor: Combining AI with no-code/low-code automation to execute tasks that otherwise consume manual effort.
  • Core capabilities of robust AI automations:
    • Understand requests (domain grounding), apply rules/logic, take actions: replies, lead qualification, CRM updates, appointment scheduling, and reporting.

What a career looks like

  • Roles/tasks include:
    • Designing AI operating systems for businesses (multi-agent and rules-driven workflows).
    • Building voice/chat agents, integrating CRMs, orchestrating lead intake and follow-ups.
    • Streamlining internal ops and reducing error rates and turnaround times.
  • Tooling commonly used: Zapier, Make.com, Voiceflow, GoHighLevel (GHL), LLMs (e.g., Claude/ChatGPT), and agent frameworks.

How to identify automation opportunities

  • Favor:
    • Find bottlenecks: repetitive, time-heavy tasks with poor ROI or high error rates.
    • Ask targeted discovery questions; use Google to research industry-specific pain points.
    • Prioritize processes where automation measurably improves efficiency or customer experience.
  • Chidera:
    • Automate repetitive, time-consuming workflows; preserve human-in-the-loop where empathy/judgment is essential.

Sustainability and market direction

  • Favor: Sustainable if you’re a problem-solver with foundations in AI behavior, workflows, and business process thinking. Continuously learn and adapt.
  • Chidera: The space moves fast, and some big platforms are embedding automation (e.g., Slack agents). To endure, broaden from “automation builder” to an “AI person” (prompting, agent design, product thinking), so you’re the go-to for AI-driven solutions—not just connectors.

Typical challenges (and how to overcome them)

  • Information overload: focus on fundamentals and expand deliberately.
  • Impostor syndrome and self-doubt: you learn fastest by building for real constraints; don’t wait to be “fully ready.”
  • Learning alone: find accountability partners or communities; build in public to sustain momentum and attract opportunities.
  • Positioning: being skilled is not enough—clearly communicate business value (problems solved, outcomes delivered) rather than tool jargon.

Speaker Journeys and Milestones (Condensed Lessons)

  • Michael Ajiboye (AI/ML → Prompt Engineering)

    • Progressed from Python and data science (pandas, seaborn, matplotlib) to ML (NLP, vision), then into LLMs and prompt evaluation.
    • Competed in hackathons (top placements) and participated in a research lab with access to frontier models; focused on mitigating hallucinations and optimizing prompts.
    • Key lesson: leverage technical grounding to extract maximum value from LLMs; be tool-agnostic but outcome-obsessed.

  • Colin Eromosele Okunowo (Creative → Prompt Engineering)

    • Painful, fragmented creative workflows led to integrating AI to speed ideation, scripting, and design.
    • Key lesson: prompt engineering spans many industries; consistency and focus breed mastery and opportunity.

  • Favor (Ops → AI Automation)

    • Identified manual lead follow-up inefficiencies; started with GHL automation; scaled into building AI operating systems.
    • Key lesson: domain insight (ops, CRM, CX) + AI/no-code tools = high-impact solutions. Financial upside was significant compared to prior roles.

  • Chidera Okonkwo (CS/UI-UX → AI Automation)

    • Discovered Zapier via research; shifted to Make.com; built voice agents with Voiceflow; first client arrived after a WhatsApp status post; later contracts via building in public.
    • Key lesson: start simple, ship publicly, and grow into more complex agentic systems; cultivate a problem-solver identity.

Human vs. Automation Boundaries

  • Automate:
    • High-frequency, repetitive, time-consuming tasks (e.g., inbound triage, appointment scheduling, standardized lead qualification, CRM updates).
  • Keep humans in the loop:
    • Low-frequency tasks, nuanced cases needing empathy/judgment, high-risk decisions.
  • Operating principle: use human-in-the-loop (HITL) checkpoints for reviews/approvals where appropriate.

Q&A Highlights

  • Getting started “with just a phone” and authentic learning

    • The bootcamp offers structured instruction (including AI automation). For deeper practice, you’ll eventually need a desktop environment to build and test workflows.

  • No technical background (e.g., medical field): where to start?

    • You can start in AI automation without prior coding. Begin with Zapier or Make.com. Leverage your domain knowledge (e.g., healthcare) to target valuable workflows.

  • Do you need data analysis for automation?

    • No. Helpful thinking skills grow through building real workflows. Use search and AI assistants to plan solutions; focus on business problems and outcomes.

  • CRM workflow automation vs AI automation

    • CRM automation is a subset within the broader AI automation landscape. Many CRMs (e.g., GHL, HubSpot) now embed AI/agent features, enabling end-to-end customer workflows.
    • Time to competence: with focused effort, ~3 months can get you functional proficiency in no-/low-code stacks.

  • Building a trading bot: automation vs ML

    • Automation and ML are distinct. ML involves model training/coding; automation orchestrates tools and logic, often minimal code.
    • You can prototype with LLM coding aids (e.g., Claude Code) if you pursue bot logic. If you want full custom ML, expect deeper coding.

  • Will AI/agents replace automation roles?

    • Some platform-native agents will absorb routine automations (e.g., Slack agents). Future-proof by expanding into prompting, agent design, process architecture, and product-oriented AI work.

Tools, Models, and Concepts Mentioned

  • No-code/low-code: Zapier, Make.com, Voiceflow, GoHighLevel.
  • LLMs/agents: ChatGPT, Claude (incl. Claude Code), LLaMA; RAG (Retrieval-Augmented Generation); hallucination mitigation; system prompts; model evaluation.
  • Data/ML foundations (from journeys): Python, pandas, seaborn, matplotlib, transformers, deep learning, NLP.
  • Creative stack touchpoints: Blender, color grading; text-to-image/video for ideation.
  • Platform-native agents: Slack agents (example of embedded enterprise automation).

Practical Takeaways for Learners

  • Pick one path first, go deep for 3–6 months: consistency beats breadth.
  • Define outcomes up front: who is the audience, what problem, what constraints, what does “good” look like?
  • Build in public: share learnings and small wins (helps accountability and attracts opportunities).
  • Create a portfolio with real use-cases: even self-initiated projects in a chosen domain.
  • Communicate value in business terms: time saved, error reduction, faster throughput, better CX—not just “I know Tool X.”
  • Expect rapid change: keep fundamentals (prompt craft, process thinking, evaluation) at the center and pivot tools as needed.
  • Overcome impostor syndrome by doing: apply early, volunteer, freelance, or tackle internal projects to accelerate learning.

What the Field Has Done for the Speakers

  • Chidera: Shifted mindset to “problem solver,” spotting automatable opportunities everywhere.
  • Michael: Gained “wings” to contribute across domains (branding, research, software, business ops) through AI.
  • Colin: Learned to see solutions/opportunities and to teach/share insights at scale.
  • Favor: Significant financial uplift and access to blue-chip and public-sector projects; reinforced that value is measured by problems solved, not number of tools known.

Program and Next Steps

  • The session is part of a broader mentorship series aligned to the bootcamp. Applications (Cohort 8) were closing within ~48 hours of the webinar date.
  • Learners were encouraged to:
    • Choose one skill track (prompt engineering or AI automation) for focused study during the bootcamp term.
    • Engage actively in community, ask questions, and prepare portfolios/capstones demonstrating real-world application.