– Your own Vantage point
– Your Context
– Where you, your team, organisation and the industry sit on the emergent process of advancing their craft
– Where you, your team and organisation sit on the AI learning curve
To identify and overcome your biggest challenge, it helps to understand why different organisations face distinct challenges and how these key factors play a role. This is covered below.
Here, you can get a sense of the challenges many others are experiencing, and realise that, nonetheless, you are not alone.
Examples of the challenges of AI-Augmented Development<\/strong><\/strong> in Software Development<\/strong><\/p>\n\n\n\n
- \n
- Codebase and SDLC AI unreadiness<\/li>\n\n\n\n
- Lack of AI skills and experience<\/li>\n\n\n\n
- Code Review Indigestion (AKA Review fatigue)<\/li>\n\n\n\n
- Cognitive Debt (AKA Systems and domain knowledge & understanding erosion)<\/li>\n\n\n\n
- AI-induced technical debt (surging incidents, a Winchester Mystery House Architecture or lack of consistency in the code)<\/li>\n\n\n\n
- Cognitive Fatigue (due to AI inhuman reasoning, interaction and presentation style, and to fluctuating non-deterministic results)<\/li>\n\n\n\n
- CI\/CD Pipeline Overload<\/li>\n\n\n\n
- Inadequate team structure and size<\/li>\n\n\n\n
- Outdated team practices, tools and ways of working<\/li>\n\n\n\n
- …<\/li>\n<\/ul>\n\n\n\n
E<\/strong>xamples of the challenges of AI-Augmented Development in Product and Scope Management<\/strong><\/p>\n\n\n\n
- \n
- Product Managers’ exhaustion (not keeping up with teams’ speed)<\/li>\n\n\n\n
- Widening Product-Tech teams distance (and friction)<\/li>\n\n\n\n
- Features factory (more features shipped with limited impact on user satisfaction, value creation for stakeholders, and learning)<\/li>\n\n\n\n
- User exhaustion (the number of product changes shipped exceeding users\/customers’ need and ability to absorb change)<\/li>\n\n\n\n
- Outdated Product and scope management practices, tools and ways of working<\/li>\n\n\n\n
- …<\/li>\n<\/ul>\n\n\n\n
Examples of the challenges of AI-Augmented Development in the Business<\/strong><\/p>\n\n\n\n
- \n
- Failed AI replacing employees initiatives instead of augmenting employees<\/li>\n\n\n\n
- Strategy, budgeting, initiative selection, and funding processes not keeping up with the Prod-Tech speed<\/li>\n\n\n\n
- Cultural friction (toward change, innovation, adaptability, speed)<\/li>\n\n\n\n
- Initiatives shipping faster without meeting more strategic value objectives<\/li>\n\n\n\n
- Growing AI LLM bills without a measurable ROI<\/li>\n\n\n\n
- Structure and roles hindering Biz-Product-Tech collaboration and productivity<\/li>\n\n\n\n
- Contracts hindering customer collaboration<\/li>\n\n\n\n
- …<\/li>\n<\/ul>\n\n\n\n
Now let’s look at the factors contributing to the biggest challenges you are facing.<\/p>\n\n\n\nVANTAGE POINTS<\/h2>\n\n\n\n
Most of the reports focus narrowly on the Software Development Life Cycle (SDLC). However, my experience, client conversations, and trusted peer insights reveal a different story. Even individuals within the same organisation may point to different primary challenges depending on their vantage point. <\/p>\n\n\n\n
1st Takeaway: to make an impact, look end-to-end, from idea to realised value<\/strong>, not only at tech practices<\/strong><\/p>\n\n\n\n
Taking this broader view requires looking into the Tech, Product, and Business side of things. It also requires looking at different levels of granularity: at team(s), departments, and at organisational level. Tooling and harness, skills and experience, and the company’s culture of change, innovation, and reinvention deserve attention too.
The acceleration brought by AI-augmented development impacts each area in different ways, exposing different challenges depending on where you stand.<\/p>\n\n\n\nCONTEXT<\/h2>\n\n\n\n
An organisation’s specific context plays a decisive role in shaping its biggest challenge in AI-augmented development<\/span>. What succeeds in one environment may fail in another. Consider key variables such as:
– the size and complexity of the codebase
– the novelty or maturity of the product
– the level of reliability required
– the risk appetite, its potential reward, or ROI
– the skills and experience in AI-Augmented Development
– …<\/p>\n\n\n\n2nd Takeaway<\/strong>: there are no one-size-fits-all playbooks<\/strong>; what worked elsewhere may not work well for you<\/strong><\/p>\n\n\n\n
For instance, early adopters of a new high-growth product might tolerate occasional defects, while rapid experimentation may deliver massive ROI. In that setting, fully automating code reviews with AI makes sense because speed outweighs perfection.<\/p>\n\n\n\n
That same approach fails for mission-critical systems like primary banking platforms or fire safety software. In high-stakes environments, human accountability and avoiding knowledge and understanding erosion must take precedence (see, for example, Kent Beck’s 3X: Explore, Expand, Extract, and the evolution axis of Wardley Maps, from Genesis to Commodity, both showing why each stage calls for different practices). Furthermore, underlying technical debt can restrict how much AI can do autonomously, as AI left unchecked tends to amplify existing quality issues.<\/p>\n\n\n\n
EMERGENT IMPROVEMENT<\/h2>\n\n\n\n
Major Tech shifts have driven every revolution and reinvention of how we work over the last decades. With AI being the latest wave. <\/p>\n\n\n\n
These revolutions and reinventions rarely follow a linear path. Instead, they emerge gradually, often with a bottom-up push.
What begins as a set of novel engineering practices quickly reshapes collaboration, product management, and leadership. Over time, these changes ripple outward to influence team structures, coordination and stewardship, organisational governance, contracting, strategic initiative selection, funding, and budgeting.<\/p>\n\n\n\nWe are now in an era of rapid change and AI evolution that is fuelling the reinvention of how we work.
Industry-wide, we are still learning, experimenting and discovering, from deliberate practice, sharing what we learn, comparing notes across different contexts, trying to understand and identify which principles are still valid and what new principles are emerging.<\/p>\n\n\n\n3rd Takeaway: improvement in AI-augmented development is a continuous emergent process sustained by direct experience and deliberate practice<\/strong><\/p>\n\n\n\n
There is a pearl of wisdom from a previous major technological shift that suggests how to pursue mastery while exploring uncharted territories, that is extremely relevant now: We are uncover-ing<\/strong><\/span> better ways of working by doing it<\/strong><\/span>.<\/p>\n\n\n\n
LEARNING<\/h2>\n\n\n\n
While the current climate of rapid innovation and job insecurity incentivises chasing the latest hype and making bold claims, sustainable progress requires gradually building the core skills and making the experience the organisation and its context require. Skipping ahead creates fragile foundations that break under pressure.<\/p>\n\n\n\n
Learning AI-augmented development requires going through the journey of writing your prompts and loops and setting up your harness.
During that journey, human-driven manual code reviews remain a necessary inspection point to spot what is working and what is not in your prompts, loops and harness, and what you need to learn and improve.
This approach also reaches the goal of adapting your prompts, loops and harness to you, your “project”, and your company context, which is essential to make things work well, as mentioned before.<\/p>\n\n\n\n4th Takeaway: you can’t skip ahead on the learning journey; you have to walk every step<\/strong><\/p>\n\n\n\n
So where you sit on the learning curve keeps shifting, the ways of working that suit you keep adapting, and so does your biggest challenge.<\/p>\n\n\n\n
SOURCES AND FURTHER READING<\/h2>\n\n\n\n
From SmHarter<\/strong><\/p>\n\n\n\n
- \n
- Maximising AI Benefits in Software Engineering: In-Depth Strategic Overview<\/a>, 23 September 2026<\/li>\n\n\n\n
- The case for an AI-Augmented Development methodology<\/a>, 16 September 2026<\/li>\n\n\n\n
- Where AI Ends and Human Value Begins<\/a>, 8 September 2026<\/li>\n\n\n\n
- AI-assisted coding: how to ensure real productivity gains?<\/a>, 1 September 2025<\/li>\n\n\n\n
- Coding with AI: 2 alternative philosophies<\/a>, 14 May 2026<\/li>\n\n\n\n
- Organisational strategy for AI-Assisted coding adoption<\/a>, 25 August 2025<\/li>\n<\/ul>\n\n\n\n
Others<\/strong><\/p>\n\n\n\n
- \n
- The AI-Native SDLC playbook<\/a>, Anthropic, 21 August 2026<\/li>\n\n\n\n
- State of AI-assisted Software Development 2025<\/a>, DORA<\/li>\n\n\n\n
- The ROI of AI-assisted Software Development<\/a>, DORA, 2026, as summarised by InfoQ<\/a><\/li>\n\n\n\n
- Finding balance in the era of tokenmaxxing<\/a>, DORA Insights, 2 June 2026<\/li>\n<\/ul>\n\n\n\n
<\/div>\n\n\n\n
\n\n\n\n\n\n![\"\"](\"https:\/\/www.smharter.com\/blog\/wp-content\/uploads\/new-philosophy.png\")
<\/figure>\n<\/div>\n\n\n\n\n\n\tNavigate Change & Innovation<\/h3>\n\t
\n\t
\n\tSee how we can help.\n\t
\n\tYou, your team, your organisation.\n\t
\n\t<\/p>\n\n\t\n\t\t\n\t\tOur Services\n\t\t<\/a> \n\t\t\n\t\tBuy the Book\n\t\t<\/a>\n\t<\/div>\n<\/div>\n<\/div>\n<\/div>\n\n\n\n<\/p>\n","protected":false},"excerpt":{"rendered":"
Coding is no longer the biggest bottleneck. So what is the biggest challenge? Discover why organisations, and even colleagues, give different answers, and why that matters for overcoming yours.<\/p>\n","protected":false},"author":2,"featured_media":4064,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[9,7],"tags":[],"class_list":["post-4055","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai-strategy","category-ai-assisted-coding"],"_links":{"self":[{"href":"https:\/\/www.smharter.com\/blog\/wp-json\/wp\/v2\/posts\/4055","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.smharter.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.smharter.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.smharter.com\/blog\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.smharter.com\/blog\/wp-json\/wp\/v2\/comments?post=4055"}],"version-history":[{"count":50,"href":"https:\/\/www.smharter.com\/blog\/wp-json\/wp\/v2\/posts\/4055\/revisions"}],"predecessor-version":[{"id":4125,"href":"https:\/\/www.smharter.com\/blog\/wp-json\/wp\/v2\/posts\/4055\/revisions\/4125"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.smharter.com\/blog\/wp-json\/wp\/v2\/media\/4064"}],"wp:attachment":[{"href":"https:\/\/www.smharter.com\/blog\/wp-json\/wp\/v2\/media?parent=4055"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.smharter.com\/blog\/wp-json\/wp\/v2\/categories?post=4055"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.smharter.com\/blog\/wp-json\/wp\/v2\/tags?post=4055"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
- State of AI-assisted Software Development 2025<\/a>, DORA<\/li>\n\n\n\n
- The case for an AI-Augmented Development methodology<\/a>, 16 September 2026<\/li>\n\n\n\n
- Maximising AI Benefits in Software Engineering: In-Depth Strategic Overview<\/a>, 23 September 2026<\/li>\n\n\n\n