Most productivity tools count hours. My AI coach recognises patterns.
The Problem
Knowledge workers with ADHD (and plenty without) know the routine: you start the day with a clear priority. Three hours later you have opened five other fronts, none of them on your list, while the actual task remains untouched.
That is not a discipline problem. That is a structure problem. And structure is something a system can provide.
“structure”, however, does not mean calendar blocking or Pomodoro timers. Those methods have been around for years, and they fail regularly for one reason: they know nothing about the person using them. They ignore energy curves. They do not recognise avoidance patterns. They fail to notice when someone has pushed the same task for three weeks, always with a plausible excuse that shifts on closer inspection.
The Approach
The coach is not an app. It is a layer in the 6-layer architecture with access to every other layer: What was planned? What was actually worked on? Which patterns recur across weeks?
The leverage comes from aggregation over time. A single unproductive day is an anecdote. Three weeks of systematically avoiding a specific task type is data. The coach tracks these time series because it runs across every session. It does not just compare plans against outcomes; it recognises the patterns behind the deviations.
Three Principles
1. Directness Over Diplomacy
No “Maybe you could consider…” Instead: “You’ve been putting this off for three weeks. Is that a conscious decision?”
A coach that only affirms is useless. The skill lies in being uncomfortable without being destructive. That works when critique rests on data, not opinions. “This week you invested 14 hours in concept work and 0 hours in outward communication” is not a judgment: it is an observation. What happens next is up to the user.
Yet directness has boundaries: acknowledge progress before pointing out the gap. That is not softening. It is context-aware feedback. Telling someone where they fall short without showing how far they have come distorts the picture.
2. Energy Before Efficiency
Not “How do I get more done?” but “How do I get the right things done with the energy I have?”
For people with chronic illnesses, parents with young children, or anyone managing finite energy, this is the core problem. Some days allow for three hours of deep work. Other days only administrative work is realistic. A good coach recognises the difference and adjusts priorities, rather than grinding through a static weekly plan.
That sounds obvious, but it changes the logic: the coach must learn that “less, but right” is often better advice than “just one more task.” Particularly with ADHD, where hyperfocus is deceptive: three hours of intense output followed predictably by a crash. Spot the pattern early, and you can intervene before the drop.
3. Patterns Over Moments
A single unproductive day means nothing. Three weeks of avoidance is a pattern. The coach aggregates over time, surfacing what the individual cannot see from the inside.
What the Coach Concretely Recognizes
Communication Debt
Borrowed from software engineering (“Technical Debt”) and applied to communication: every delayed reply, every avoided conversation, every postponed check-in is debt that accrues interest. The coach detects this accumulation, not by scanning email text, but by watching specific task types systematically slip down the backlog.
Take a channel left untouched for weeks, despite visibility being an explicit strategic goal. The coach notes the imbalance: concept work (intrinsically rewarding) dominates; communication (extrinsically driven, high friction) is avoided. The intervention is not “You need to post more,” but: “This channel has been silent for 24 days. A short post takes 20 minutes. What’s holding you back?”
The question “What’s holding you back?” matters. It creates space to reflect rather than handing down instructions.
Avoidance Patterns
Certain tasks stall repeatedly. Not because they are trivial, but because they are emotionally draining, bureaucratic, or dull. The coach identifies the pattern and targets the underlying block.
A standard ADHD failure mode: the task is not too big. It is too unstructured. The missing element is not motivation, but a concrete first step. “Do the tax return” paralyses. “Open the folder and collect the wage tax certificate” works.
Over-Conceptualizing Instead of Finishing
A common trap for analytical minds: endless refinement. Version 1, version 2, version 3, all kept internal, while no one has ever seen version 1. The coach asks the inconvenient question: “Is this revision better for the recipient, or just for you?”
This is not a critique of quality standards. It is the observation that perfectionism and avoidance are closely linked. Endless optimisation shields you from external scrutiny.
Energy Cycles
When does deep work actually happen? When is capacity limited to admin? Over several weeks, the coach models the individual’s capacity curve. Not as a rigid scheme, but as an empirical tendency: “Your pattern shows that late-night sessions are productive, but a price is paid the next morning.”
Making Invisible Progress Visible
People with ADHD systematically discount their own output. Attention stays fixed on what remains undone. Here, the coach steps in: “In the last two weeks, you built three systems, published a podcast, and submitted a strategy. That’s not nothing.” No flattery. An enumeration of facts.
The Psychological Debate: Does the Person Lose Their Autonomy?
This question surfaces in every serious discussion of AI coaching, and it warrants a direct answer.
The argument against: If a system directs my schedule, flags my avoidance loops, and tracks my energy curves, do I lose the ability to self-reflect? Does self-awareness atrophy once outsourced?
This is not a straw man. It is a genuine risk.
The argument for: The assumption that humans excel at unassisted self-reflection is empirically weak. Kahneman documented at length how systematically we misjudge our own cognitive processes. Therapists exist because introspection alone regularly fails. Coaching exists because an external vantage point captures what internal perception systematically misses.
My position: The coach does not replace the capacity for self-reflection. It extends it. It surfaces data no individual practitioner could assemble, simply because no single observer holds all the pieces. A therapist gets one hour a week. A colleague sees only professional output. A partner sees only private life. The coach spans both environments, tracking how they influence one another.
This is not surveillance. It is a mirror sharper than unassisted self-perception.
That holds only if the human keeps final authority. The coach recommends; it does not decide. It highlights patterns; it does not act. Above all, it must remain fallible and open to correction. A coach that cannot be corrected is not a coach. It is a dictator.
What the Coach Is Not
Not a therapist. Not a time-tracking tool. Not a surveillance system. The coach cannot inspect unshared content. It observes patterns, not thoughts.
It does not replace medical care or pharmacology. It is one instrument within a broader support structure. Its utility lies precisely in the gap between weekly therapy sessions (one hour) and daily operational reality (the remaining 167 hours).
It also makes mistakes, regularly. Not every flagged “pattern” holds up. Sometimes a task sits untouched for three weeks simply because it does not matter right now. Accepting corrections and adjusting its baseline is built into the architecture.
The Connection to the Self-Vector
The coach is the initial application of the Selbstvektor (self-vector). Once a system maintains a structured model of the user (working habits, energy curves, communication preferences, motivational triggers), it can anticipate rather than merely react.
“You did three hours of deep work this morning. Your pattern shows that after 3 PM only administrative tasks make sense. Should I reorder the priorities?” That is not science fiction. That is a weighting function applied to existing operational data.
Here lies the fundamental difference from conventional productivity tooling: the coach does not react to manual inputs. It reacts to a continuous model of the user. That model sharpens with every interaction: not by accumulating arbitrary data, but by refining its interpretation of the data it already holds.