Cognitive Peptides for Founders Who Don't Want a Modafinil Habit: Selank, Semax, and What the Russian Literature Actually Shows

Every founder I know has spent some amount of time in the modafinil funnel. The drug works. It produces wakefulness, narrows attention, and gets you through a red-eye or a crunch quarter you would otherwise lose. It also requires a prescription in most jurisdictions, sits in a controlled-substance category that operators running a serious health stack tend to scrutinize before committing, and produces tolerance in a meaningful subset of users who run it regularly.

When founders start asking whether there is an alternative — not more caffeine, not a stimulant with a different brand name, something mechanistically different — they end up on the Russian peptide literature. Selank and Semax. Developed at the Institute of Molecular Genetics in Moscow. Administered intranasally. Found, usually, through a nootropic blog post that does not read the studies very carefully.

This issue is the founder-grade read. It works through an 8-week intranasal protocol as a worked methodological example — Selank in the AM, Semax pre-deep-work, tracked the way Issue 4's framework requires — and reads the Russian literature as carefully as the translation and the publication record allows. The evidence is better than the nootropic blog coverage implies. The healthy-volunteer extrapolation is not established. Both of those things are true at once, and that tension is the whole point of this issue.

Issues 5 and 7 were the previous two application-arc issues — the Wolverine stack and the Growth Stack. This is the third and final application-arc issue before the Edge arc begins. The Issue 3 half-life nuance applies here in a specific way I will flag when we get to the mechanism.

Selank. A synthetic heptapeptide analog of tuftsin, an endogenous tetrapeptide that occurs naturally in the body. Developed at the Institute of Molecular Genetics and the V.V. Zakusov Institute of Pharmacology in Moscow, with publication activity concentrated in the 1990s through 2010s. The primary mechanism in the literature is anxiolytic via GABAergic modulation — specifically, enhancement of GABA-A receptor sensitivity — paired with BDNF elevation in rodent models and reported effects on enkephalin and endorphin pathways. Delivered intranasally. The plasma half-life of the parent compound is short — roughly 2–3 minutes — but the CNS effects persist longer than that figure implies. This is the same Issue 3 nuance: plasma half-life and CNS-effect duration are not the same number, and conflating them produces incorrect intuitions about dosing schedule. The Russian clinical literature does not associate Selank with tolerance development.

Semax. A synthetic analog of the N-terminal fragment of ACTH (adrenocorticotropic hormone) — specifically ACTH(4–10). Also developed at the Institute of Molecular Genetics. The most consistent finding across the published literature is BDNF upregulation. Secondary mechanisms include enhanced norepinephrine and dopamine turnover in the prefrontal cortex, and neuroprotective effects in ischemia and hypoxia models. Also administered intranasally. The NE/DA prefrontal-cortex mechanism is what operators recognize as pre-deep-work sharpness. The important gap: the literature tested Semax in clinical populations — ischemia patients, neurorecovery subjects — not in healthy cognitive-optimization populations. That gap matters for every inference in Section 2.

The honest summary is this: the evidence is real, and the population selection limits how far you can carry it.

Selank has the stronger anxiolytic signal. Multiple Russian RCTs (Tier 1, small — n=30–80 range) in generalized anxiety and mixed neurotic disorders show meaningful response versus placebo on standard anxiety scales, including the Hamilton Anxiety Rating Scale and the HADS. The effect size is real. The therapeutic indication in that literature is anxiety-adjacent, not cognitive-enhancement-adjacent. A founder running Selank for focus is using an anxiolytic, not a nootropic. That distinction matters because the mechanism tells you what kind of improvement to expect: a subtraction — less background noise — rather than an addition — more stimulant throughput.

Semax has the stronger cognitive signal, but in impaired-baseline populations. The ischemia and stroke-recovery literature is deep by Russian clinical standards — multiple studies, reproducible BDNF signal, observer-reported improvements in attention, memory, and processing speed in patients recovering from ischemic events. The BDNF upregulation mechanism is the most translatable hypothesis to a healthy-population cognitive-optimization use case. The controlled healthy-volunteer data, however, is thin. The jump from ischemia-recovery data (Tier 2) to a prediction about a healthy founder at a standing desk is a Tier 4 inference. Name it as such.

What the nootropic blog coverage consistently gets wrong: it treats the ischemia literature as if it were a healthy-user cognitive-enhancement trial. The evidence is real. The population generalizability is not established. Those are different statements, and conflating them is how an operator ends up with misplaced confidence in a compound they do not fully understand.

A clean methodological walkthrough of this kind of self-trial, applying Issue 4's framework — pre-protocol, single-variable rule, blinded subjective tracking, post-labs — would track the following.

Pre-protocol. Subjective focus quality is rated on three axes for two weeks before introducing any new compound — speed-to-deep-focus (minutes of friction before the work feels locked in), sustained-attention window (minutes before the first distraction pull), and end-of-session depletion (1-to-10, lower is better). Logged in a plain text file before coffee each morning, specifically to keep the caffeine signal out of the baseline. The morning window before coffee produces the cleanest focus-quality read because there is no stimulant already running. A CBC, CMP, and CRP panel is pulled the week before — neither Selank nor Semax is known to produce measurable bloodwork changes, but the pre-labs are the methodological floor Issue 4 requires.

Single-variable rule. Caffeine intake, training volume, sleep target, and all other supplementation are held constant for the full eight weeks. The new inputs in the worked example are Selank and Semax delivered intranasally — the route the Russian clinical literature and community discussion both describe — with Selank framed as an AM input and Semax framed as a pre-deep-work input, the timing pattern that follows from the mechanism descriptions in Section 1.

What this kind of self-trial would track.

Weeks 1–2: Minimal effect on the focus axis would be expected from the published mechanism. An anxiolytic effect from Selank — quieter background noise, less low-grade morning restlessness preceding deep work — could surface by mid-first-week if the GABA-A literature translates. This is the subtraction-not-addition signal: Selank removes something rather than adding throughput. A stimulant adds. The removal is what would change the output.

Weeks 3–6: Speed-to-deep-focus would be the axis most likely to move on the log if the Semax NE/DA prefrontal mechanism translates. A friction-window drop from baseline is the expected directional signal. Sustained-attention window typically shows smaller movement. End-of-session depletion is the secondary axis to watch.

Weeks 7–8: Stability check. No tolerance signal in either compound's published literature, but the protocol is the place to confirm that in any specific biology.

Post-protocol: CBC, CMP, CRP rechecked against pre-labs.

The honest read on a stack design like this. Running both compounds together prevents isolating the Selank anxiolytic effect from the Semax cognitive effect. Issue 4's single-variable rule would require running each compound alone for four weeks before combining them. A cleaner replication of this kind of self-trial would isolate one compound at a time before combining. The data generated under that sequencing is more useful than data from a parallel start.

The subtraction Selank is described as producing — an anxiolytic floor — may be the more operationally useful finding in this kind of trial. Modafinil is an addition. Selank is a subtraction. Both move the output metric. They do it through entirely different mechanisms, and a founder who understands that distinction is choosing deliberately rather than defaulting to the highest-stimulant option.

The honest tradeoff, without smoothing it over.

Modafinil: Tier 1 human data in healthy volunteers and shift-work populations. Well-characterized mechanism — wakefulness promotion via orexin-system interaction and dopamine reuptake inhibition. Prescription required. Controlled-substance classification in most countries. Tolerance reported in a subset of regular users. The deepest human safety record in the wakefulness-promotion category.

Selank + Semax: Tier 1 in anxious-population studies for Selank; Tier 2 in impaired-baseline populations for Semax; Tier 3 mechanistic for both; Tier 4 for healthy-volunteer cognitive-enhancement extrapolation. No prescription required in an RUO context. No tolerance signal in the published literature for either compound. Shorter published history with healthy cognitive-optimization operators.

A founder who does not want modafinil's tolerance profile and is comfortable operating at a lower evidence tier makes a coherent choice. A founder who expects modafinil's Tier 1 healthy-volunteer certainty from Selank and Semax will be disappointed. The evidence tier is not a disqualifier. It is the information required to choose deliberately.

Selank is an anxiolytic that produces a subtraction-based focus improvement — less background noise, lower friction to deep work. Semax is a BDNF-mechanism cognitive-support compound with a strong ischemia literature and thin healthy-volunteer data. Together they are a coherent non-stimulant stack for the operator who knows the evidence tier they are working in. The Russian literature is better than the nootropic blog coverage implies. It is also not a substitute for what does not yet exist: controlled healthy-volunteer trials in English that directly test the cognitive-enhancement hypothesis. Work with what the literature actually shows, not what the community has decided it shows.

Issue 09 — Everyone Is Wrong About Retatrutide Bloat: It's Not the Reta, It's the Titration. The contrarian read is in. Community reports, mechanism, and why the bloat complaint is a titration-speed artifact, not a compound problem.

Colby

For research use only. Not medical advice. Compounds discussed are sold for research purposes; nothing here is a recommendation to use them on humans or animals. The publisher of The Compound has a financial interest in heroxbio.com, an RUO peptide vendor; this relationship may influence which compounds are covered. See the About page for full disclosure.