Post-injury programming is where client context matters most. Generic programs aren't just ineffective here — they're potentially harmful. This example shows how an AI tool approaches a return-to-training case when it has access to detailed client history and a trainer whose profile reflects experience with this population.
The trainer profile
The trainer in this example has eight years of experience, with a significant portion of their client base coming through physiotherapy referrals. Their programming philosophy prioritizes movement quality assessment in the first training block with any post-injury client, uses conservative loading that prioritizes RPE over percentage-based loading for the first four to six weeks, and avoids any movement that replicates the mechanism of the original injury during the return-to-training phase. They prefer to build volume before intensity in the early weeks of a return block and include explicit capacity-building work for the injured area alongside general training.
The client profile
The client is a 41-year-old man, intermediate training age with five years of consistent resistance training prior to injury. He sustained a grade two hamstring strain during a recreational football match seven weeks ago. He has been cleared for return to full training by his physiotherapist with a recommendation to avoid maximal sprint effort and high-speed hamstring loading for a further six weeks. His training goal is to return to pre-injury strength levels and eventually return to recreational sport. He trains four days per week with full equipment access. He has no other injury history of note.
The generated approach: weeks one and two
The AI generates a two-week introductory phase rather than jumping directly to a standard training block. The rationale embedded in the output: the client has been away from full training for seven weeks and while cleared to train, has no recent data on current strength levels. Week one is explicitly framed as an assessment and reintroduction phase.
Hip hinge patterns are included from day one but programmed conservatively — a trap bar deadlift at perceived exertion of five to six on a ten-point scale, three sets of five, with a note in the session that load should be selected to feel easy rather than challenging. The rationale is that this re-establishes the pattern and generates a baseline loading data point rather than being a stimulus for adaptation.
Hip flexion under load is limited in week one. No leg press, no deep squat variation, no lunge variation that places significant hip flexor demand. The AI generated this constraint from the injury note in the client profile and the trainer's stated approach to return-to-training programming for lower limb injuries.
Capacity-building work for the hamstring is included as an accessory in every lower body session: a Nordic hamstring curl regression (a slow, controlled lowering from a kneeling position with assistance), two sets of six, noted as a long-term hamstring tendon and muscle capacity builder consistent with current return-to-sport literature.
Programming rationale
The key programming decisions in this plan all trace back to client context. The conservative RPE targets reflect the trainer's established approach for post-injury clients and the gap in the client's recent training history. The hip hinge inclusion at low intensity reflects the trainer's preference for keeping patterns trained during return phases rather than avoiding them entirely. The hamstring capacity work reflects both the injury type and the client's stated goal of returning to recreational sport.
Without the client's injury history, clearance notes, and the trainer's post-rehabilitation programming preferences, none of these decisions would appear in a generated plan. This is the direct illustration of why client context and trainer methodology are the determinants of whether AI output is professionally usable.
What changes in weeks three through six
The plan generated for the subsequent weeks shows a systematic progression: RPE targets increase, hip-dominant loading volume builds, the Nordic regression progresses to a more demanding variation, and sprint-mechanical preparation work is introduced in week five in anticipation of the six-week restriction lifting. The AI tracks the constraint timeline from the clearance note and builds the progression around it without being prompted to do so.