When You Can't Keep Up: A Stance Guide for Engineers — Integrating Growth Mindset, Self-Compassion, and Encouragement

• Intended readers: IT engineers of any age who feel “I can’t chase new tech anymore” or “I’m falling behind”
• Prerequisites: None
• Reading time: 18 minutes

Overview

“I can’t keep up with new tech anymore.” “My peers (or junior teammates) absorb things so much faster than I do.” If you’ve ever caught yourself thinking like this, you’re not alone — most engineers have.

This anxiety used to be framed as a mid-career problem, captured in Japan’s “IT engineers retire at 35” folklore (a piece of industry mythology popularized in the 2000s). But the demographics have already moved on. According to Japan’s Ministry of Economy, Trade and Industry, the share of IT workers aged 50+ doubled from 11% in 2010 to 22% in 2020, and the average age of system engineers reached 41.8¹. The implicit “retirement” never really existed as a structural reality. And yet the feeling of falling behind hasn’t gone away — in the AI era, it has spread across generations. The Stack Overflow Developer Survey 2024 found that only 20.2% of professional developers describe themselves as “happy” at work; 32.1% are “unhappy” and 47.7% are “complacent” (going through the motions without much joy)². Only one in five is genuinely enjoying the work. A second-year engineer who feels their stack is “already pre-ChatGPT obsolete,” and a fifty-something who can’t follow the new framework study group, are facing the same structural problem.

The “retirement age” narrative is dead, but the anxiety remains because the fear of not being able to keep learning isn’t tied to age — it’s tied to structure. Stacking technical books on top of that fear doesn’t fix it. Psychology offers a way to rebuild from the stance side instead. This article integrates three frames: growth mindset (the belief that ability can be developed through effort, strategy, and help — Carol Dweck)³⁴; self-compassion (treating yourself as you would a friend rather than a critic — Kristin Neff)⁵; and evidence-based encouragement (focusing on process and potential — Y. Joel Wong’s Tripartite Encouragement Model)⁶.

Each is powerful on its own, but each also has a failure mode. Growth mindset breaks under self-criticism. Self-compassion drifts without direction. Encouragement doesn’t land without a foundation. Combined as a loop, they cover each other’s gaps. This article lays out the evidence for each, then translates the loop into concrete situations: code review failures, AI-era catch-up, and 1on1 conversations. It’s a stance blueprint for staying engaged in the field for the long haul, regardless of age.

1. The “Can’t Keep Up” Anxiety — A Cross-Generational Phenomenon

1.1 The “35 retirement” myth today (Japan context)

Drawing on government data, paiza summarizes the historical reasoning behind the “IT engineers retire at 35” myth around three factors: (1) the rapid obsolescence of the era’s frameworks and languages, (2) wage inflation from seniority-based pay, and (3) brutal long hours that favored younger workers¹. Each was a reason organizations might decide that older engineers “weren’t worth it.”

The 2010s changed that picture significantly:

• Age distribution: METI data shows the share of IT workers aged 50+ doubled from 11% (2010) to 22% (2020)¹
• Average age: Japan’s MHLW data puts the average age of infrastructure engineers and PMs at 41.8¹
• Talent shortage: METI projects an IT talent gap of 400,000 to 800,000 by 2030¹

Remote work reduced the physical-stamina dimension, and severe shortages made age-based exclusion economically unrealistic. “Move people off the floor at 35” no longer pencils out as a workforce strategy. (The exact numbers are Japan-specific, but tech workforce aging and shortages are global trends.)

That said: the death of the myth and the validity of “something shifts mid-career” are separate questions. Cognitive science shows different abilities peak at very different ages — processing speed peaks in the late teens, working memory around 30; both decline gradually after that. Vocabulary, in contrast, keeps growing into the 50s and 60s⁷, and social cognition like emotion recognition continues to mature into middle age. Mid-career isn’t a cliff. It’s the inflection point where “competing on speed” gives way to “competing on judgment and experience.” The myth’s usable kernel — you can’t keep fighting with the same playbook forever — still holds.

1.2 Early catch-up burnout in younger engineers

Since around 2020, a structurally identical anxiety has spread to younger engineers. The “AI-native” cohort — engineers two or three years out of school — increasingly say things like “the framework I learned in college is already old” or “I don’t know where I stand in a post-ChatGPT world.”

The data backs the structure. Stack Overflow’s 2024 survey put professional developers at 20.2% happy, 32.1% unhappy, and 47.7% complacent (not actively enjoying it, not quite ready to quit either)². Workplace AI tool adoption roughly doubled year-over-year (15.7% → 32.4%), so the ground is shifting fast underneath everyone².

The survey isn’t broken out by age, but that’s the point — this anxiety isn’t age-bound. If anything, juniors are getting dropped straight into “your stack obsolesces from day one” mode, hitting the wall well before 35.

1.3 The common structure

The mid-career engineer’s “I can’t follow the new tech anymore” and the junior’s “what I learned is already obsolete” start from different places but end up at the same problem: the inability to maintain an internal image of yourself as someone who keeps learning.

This is the article’s core. For both groups, the question isn’t “stay or leave the field” — it’s “how do you keep updating your role and your way of learning, and what kind of stance supports that update?” The three frames here address that supporting stance. Read this as a prescription for a cross-generational problem, not an age-specific one.

2. Growth Mindset — Reframing “Decline” as “More Learning Ahead”

2.1 Definition and caveats

Growth mindset, coined by Stanford’s Carol Dweck, is the belief that ability can be developed through effort, appropriate strategy, and help from others. Its opposite, the fixed mindset, holds that ability is mostly innate³⁴.

Dweck herself has repeatedly pushed back, in her HBR pieces, against the lazy reading that “growth mindset = effort solves everything”³. Everyone holds both mindsets situationally, and the framework was widely diluted into “just praise effort,” which produced disappointing results in the field.

The accurate version: believing ability isn’t fixed only works when paired with concrete learning behaviors — effort + strategy + asking for help.

2.2 Evidence that it works across ages

People assume mindset interventions are for kids, but Sheffler et al. (2022) studied cognitive training with older adults and found that participants with stronger pre-intervention growth mindset showed larger cognitive gains afterward⁸. The authors concluded that growth mindset may support a positive learning cycle for cognitive gains in later life.

Sample sizes are small (Study 1: n=15, Study 2: n=27), so this isn’t definitive on its own. But it does undercut the naive assumption that learning interventions just stop working as you age.

For engineers — and treating this as suggestive rather than direct application, since the study was on older adults — the implication is straightforward: operating from “my peak is behind me” causes learning behavior to contract to match. Believing “I can keep growing if I change how I learn” and continuing to invest produces results that match that belief. The interaction is age-independent.

2.3 Why growth mindset alone breaks

Growth mindset has a failure mode. People who genuinely believe “I can develop this” tend, when they hit a wall on something new, to swing toward “I’m not trying hard enough.” The belief gets fused with high standards and becomes fuel for self-criticism.

That’s where self-compassion comes in.

3. Self-Compassion — Stop the Self-Criticism Loop, Run Longer

3.1 Definition

Self-compassion, formalized by Kristin Neff at UT Austin, has three components⁵:

1. Self-kindness: When suffering or failing, treat yourself warmly rather than harshly judging yourself
2. Common humanity: See your suffering as something all humans experience, not a personal defect
3. Mindfulness: Observe difficult emotions with balanced awareness — neither suppressing nor amplifying them

This isn’t “going easy on yourself.” It’s acknowledging your inadequacies without using them as ammunition against yourself.

3.2 The “you’ll stop trying” objection doesn’t hold up

A common skepticism: “If I’m kind to myself, I’ll stop pushing.” Neff’s 2023 Annual Review of Psychology article shows the empirical evidence doesn’t support this concern⁵. People high in self-compassion actually tend to:

• Engage in more adaptive coping after failure
• Set learning goals oriented toward mastery rather than avoidance
• Show lower burnout and higher long-term job satisfaction

In short, “I have to beat myself up to perform” is an introspective illusion. People with higher self-compassion tend to push longer and more persistently⁵.

3.3 Why this fits engineering specifically

Engineering work — at every level — has three persistent features: comparison is easy, failures are visible, and obsolescence is constant. That’s a structure that breeds the self-talk of “I’m behind” and “it’s too late for me.”

Left untreated, that self-talk produces retreat from new challenges — a defensive reaction of “if I’d just get hurt anyway, why try.” Self-compassion dissolves the defense and rebuilds the foundation for trying again.

If growth mindset is the belief “I can keep learning,” self-compassion is the stance “I can forgive myself for being mid-process.” Without the latter, the former leads to burnout. Without the former, the latter leads to stagnation.

3.4 Why self-compassion alone drifts

Self-compassion’s failure mode: being kind to yourself doesn’t tell you where to step next. Tolerating stagnation and ignoring it look identical from inside. You need directional encouragement.

4. Evidence-Based Encouragement — Focus on Process and Potential

4.1 Wong’s Tripartite Encouragement Model

Encouragement traces back through Adlerian psychology, but the conceptual integration arrived late. Y. Joel Wong (Indiana University) synthesized it in a 2015 Counseling Psychologist review as the Tripartite Encouragement Model (TEM)⁶.

TEM defines encouragement as an interpersonal influence process with three elements:

• Feature: Not praise — expression that focuses on strengths, potential, and process of effort
• Goal: Evoking courage, persistence, and hope in the recipient
• Context: Difficulty, challenge, or hardship — the situations where encouragement carries meaning

The key distinction is what you focus on. Praise targets results or ability (“you’re talented”). Encouragement targets process and potential (“the way you stuck with that means you’ll find another approach next time”)⁶.

4.2 Self-encouragement also works

Wong frames TEM as interpersonal but notes it extends to self-encouragement. The application is straightforward: switch your self-talk from “result evaluation” to “process evaluation.”

The connection to growth mindset research is direct. Dweck’s prescription to “praise effort and strategy, not ability” is itself a form of encouragement-style feedback³.

4.3 Caveats

Wong notes in his review that because encouragement was integrated as a concept relatively late, the empirical base is thinner than for self-compassion or growth mindset⁶. Intervention evidence is suggestive in places. This article doesn’t treat encouragement as a panacea — it positions it as the “directional” element in a three-part loop.

5. Integration — The Learning Loop

5.1 Why integrate

Each frame has a failure mode when used alone:

• Growth mindset alone: “I can develop this” curdles into self-criticism
• Self-compassion alone: kind to yourself, but no direction emerges
• Encouragement alone: provides direction, but doesn’t land without an underlying belief and self-acceptance

Combining them as a cycle means each one covers the others’ weaknesses.

5.2 The loop

flowchart TB
    A["Challenge / new tech<br/>(field reality, age-agnostic)"] --> B["Growth Mindset<br/>Believe 'I can keep<br/>learning' and step in"]
    B --> C["Failure, plateau,<br/>not understanding"]
    C --> D["Self-Compassion<br/>Stop the self-criticism loop<br/>Forgive the mid-process self"]
    D --> E["Evidence-based<br/>Encouragement<br/>Refocus on process<br/>and potential"]
    E --> F["Choose the next<br/>small step"]
    F --> A

The pivot point is where you return after failure. The fixed-mindset + self-criticism loop returns to “I can’t do this.” The three-part loop returns to “the next small step.” The same failure produces different next actions.

6. Field Practices — Three Common Scenarios

6.1 When a code review comment lands hard

Typical reaction: “My code is sloppy.” “I have no taste.” “I did it again.” Spiral into self-criticism. Get defensive on the next PR, or quietly cut your PR frequency.

Run the loop:

• Growth mindset: “This is a learning opportunity for design skill.” (Code quality is not your worth as a person.)
• Self-compassion: “Getting hit by a review stings — that’s normal. Every engineer goes through this.” (Common humanity.)
• Encouragement: Tell yourself, “The focus to absorb that feedback this carefully is a strength. Next time I’ll run a self-review with that lens before pushing.” Process-focused.

6.2 When new AI tools and frameworks won’t stop appearing

Typical reaction: “I can’t track all of this.” “Juniors absorb it faster.” Surface-skim everything, internalize nothing, get more anxious.

Run the loop:

• Growth mindset: “I might lose on speed, but I can compensate with depth. Time to update how I learn.”
• Self-compassion: “Not tracking everything is an information-volume problem, not a personal-ability problem.”
• Encouragement: “Narrowing it down to the two tools that actually move my work — that’s judgment.” Process recognized.

6.3 When you’re in a 1on1 with a report or junior

This is where encouragement does its original interpersonal work.

• Result-evaluation feedback: “The release was high quality.”
• Encouragement-style feedback: “I noticed you reworked the design multiple times to hold the quality bar even with that schedule. That kind of persistence is going to be a strength on any complex project ahead.”

The latter focuses on the person’s process and potential⁶. Teams that habituate this style tend to free members from the “if I don’t ship results I won’t be valued” fear, which raises long-term engagement.

The critical caveat: encouragement is not flattery. Excessive praise that isn’t grounded in observation doesn’t qualify as encouragement under Wong’s definition. Encouragement is restating observed facts in the language of process and potential.

Application to “the era of leaders who can’t teach”

Encouragement does another piece of heavy lifting in modern 1on1s — managing in the era where you can’t teach the technical content yourself.

The pace of technical change has reached the point where leaders often can’t directly teach the technology to their reports. That’s a structural condition, not a leader-capability problem. But when teaching becomes impossible, feedback tends to drift toward “mindset,” “attitude,” and “stance” — and feedback in that direction crosses into harassment territory the moment it overshoots. “Your mindset isn’t there yet” leaves the recipient nowhere to escape, unlike technical feedback.

What works instead is for the leader to dissolve the rigid “teacher / learner” structure and stand alongside the report as a co-investigator. Return encouragement-style feedback on the report’s process of trial and error, while showing — through behavior, not words — that you’re applying growth mindset to yourself too: “I don’t know either, let’s dig in together.”

From the report’s perspective, this stops being “evaluated by someone who has the answer” and becomes “facing the same unknown together.” The three-part loop runs not just within an individual but between people in pairs and teams.

7. Pitfalls — Use Self-Compassion on the Pressure to “Grow”

A common trap once you learn these three frames: turning the concepts themselves into evaluation tools. “I need to develop a growth mindset.” “I’m not doing self-compassion right.” Now the framework is a new measuring stick for your inadequacy.

This is just the self-criticism loop running on a new substrate. As Neff repeatedly notes, the moment you start grading yourself on whether you’re “doing self-compassion well,” it’s no longer self-compassion⁵.

The fix is mechanical: apply the three elements to “the version of yourself that can’t apply the three elements.” Talk to yourself like this: “My mindset wobbled today. That’s normal — every engineer has those days (common humanity). The fact that I’m still trying to stay aware of this is itself progress in process terms (encouragement). I can wobble and still come back (growth mindset).” If you have the loop as a structure, don’t try to stand outside the loop — just get back on it.

A note for readers who want to go deeper on the self-talk side: “observing negative self-talk instead of fighting it” is the core stance of ACT (Acceptance and Commitment Therapy). This article addresses the underlying stance; ACT provides specific self-talk-level techniques. They complement each other. See the related article below if you want to dig into the practice side.

8. Individual Effort vs. Organizational Environment

One important caveat to close on. This article is not a “fix your stance and your career anxiety disappears” individual-responsibility argument.

If your organization excludes people based on age or tenure, applies excessive short-term performance pressure, or fails to carve out time for learning, individual mindset can only go so far. The self-compassion research above shows reduced burnout — but only “within the same environment.” Changing the structures that generate burnout is a different layer of work⁵.

The healthy version: individuals develop the three elements, and organizations adopt encouragement-style feedback as a team norm. Both run in parallel. The 1on1 in section 6.3 is the meeting point. If your team operates on a “result evaluation + character evaluation” culture, members will struggle to maintain self-compassion. If the team norm is encouragement-style, individuals’ three-part loops are reinforced.

This article covers the individual side. The companion question — “how to make encouragement a team norm” — is the organizational side, and remains as a separate topic.

Conclusion

The “retirement at 35” folklore is dying as a description, but its embedded warning — you can’t keep fighting with the same playbook — is alive. In the AI era, that warning has reached not just mid-career engineers but also juniors. The cross-generational problem is continuously updating your role and your way of learning.

This article integrates three psychological frames as the stance that supports that update:

• Growth mindset: Believe ability develops through effort and strategy (without misreading it as “effort solves everything”)
• Self-compassion: Treat the in-progress version of yourself as a friend, not a critic
• Evidence-based encouragement: Focus on process and potential, not results — for yourself and for others

Each has a failure mode alone. Combined as a loop — failure → re-orientation → next step — they reinforce each other. Translated into code reviews, new-tech catch-up, and 1on1 conversations, the loop lets you treat your current position as “the start of the next stage” instead of “peak passed” or “the end.”

This isn’t a blueprint for crushing it in business. It’s a stance design for staying engaged in the field for the long haul, regardless of age. Don’t try to master it perfectly. Weave it into your day as part of the loop. That’s probably what makes it last.

Related Articles

• When Negative Self-Talk Won’t Stop — What ACT Changes Instead of “Stopping Thoughts” — Concrete techniques (acceptance, cognitive defusion) for the “I’m falling behind” self-talk
• The Science of Aging and Cognition: What Declines, What Grows — Age-by-age cognitive change, scientifically organized
• The Psychology of Perfectionism: Where High Standards Tip into Self-Destruction — Self-compassion’s relationship to perfectionism
• The Programmer Obsolescence Debate — A Shift in How to Learn — Learning stance for the AI era
• Define Your Career Plan by Contribution, Not Skills — Limits of skill-centric career thinking

References

References are numbered to match the inline citations.

1. Is the “IT engineers retire at 35” myth a thing of the past? Age distribution and salary breakdown - paiza Development Blog (2025). Cites METI and MHLW (Japan) data. Reliability: Medium (industry media citing public data) ↩︎ ↩︎² ↩︎³ ↩︎⁴ ↩︎⁵

2. Stack Overflow Developer Survey 2024: Professional Developers - Stack Overflow (2024). Pro developer job satisfaction (happy 20.2% / complacent 47.7% / unhappy 32.1%); workplace AI tool adoption doubling (15.7% → 32.4%). Reliability: Medium-High (large-scale but self-selected sample) ↩︎ ↩︎² ↩︎³

3. What Having a “Growth Mindset” Actually Means - Carol S. Dweck, Harvard Business Review (2016). Reliability: Medium-High (commentary by the originator) ↩︎ ↩︎² ↩︎³ ↩︎⁴

4. Dweck, C. S. Mindset: The New Psychology of Success. Random House (2006/2016). The foundational monograph for the growth mindset concept. Reliability: High (foundational text) ↩︎ ↩︎²

5. Self-Compassion: Theory, Method, Research, and Intervention - Kristin D. Neff, Annual Review of Psychology, Vol. 74 (2023). Reliability: High (peer-reviewed review) ↩︎ ↩︎² ↩︎³ ↩︎⁴ ↩︎⁵ ↩︎⁶

6. The Psychology of Encouragement: Theory, Research, and Applications - Y. Joel Wong, The Counseling Psychologist, 43(2), 178–216 (2015). The Tripartite Encouragement Model. Reliability: High (peer-reviewed) ↩︎ ↩︎² ↩︎³ ↩︎⁴ ↩︎⁵

7. When does cognitive functioning peak? The asynchronous rise and fall of different cognitive abilities across the life span - Joshua K. Hartshorne & Laura T. Germine, Psychological Science, 26(4), 433–443 (2015). Reliability: High (peer-reviewed, large sample) ↩︎

8. Growth Mindset Predicts Cognitive Gains in an Older Adult Cognitive Training Program - Pamela Sheffler et al., International Journal of Aging and Human Development (2022). Reliability: Medium-High (peer-reviewed but small samples: Study 1 n=15, Study 2 n=27) ↩︎