The short answer: Hire the mix engineer whose existing work sounds like the record you're trying to make, who will put deliverables, revisions, and turnaround in writing before money moves, and who asks you smart questions about the song before quoting anything. Skip anyone who promises a hit, refuses revisions, won't name a file format, or runs the same preset chain on every song regardless of genre. Everything else — gear lists, studio photos, plugin brand loyalty — is decoration.

Mixing is a service business built on taste and communication, and most bad experiences come from the business half, not the taste half. Here's how to vet properly.

Start with discography relevance, not discography size

A long credit list means someone has been working a while. It doesn't mean they can mix your song. What you actually want is relevance: has this person mixed records in your lane, at the level you're aiming for, recently?

For R&B, hip-hop, and afrobeats specifically, that means listening for things a rock or pop-heavy mixer often gets wrong:

  • Sub and 808 discipline. Does the low end translate on a phone and a car? Can you still hear pitch movement in the 808 at 40-60 Hz, or is it a blurry thud? Are the kick and 808 sharing space through ducking or careful arrangement, not just a low-cut?
  • Vocal presence without harshness. Modern R&B and rap vocals sit forward and loud. Listen around 2-5 kHz for edge and around 6-9 kHz for sibilance. A good mixer gets intelligibility without making you flinch on cheap earbuds.
  • Ad-lib and stack architecture. In rap and R&B, the doubles, ad-libs, and harmony stacks are half the arrangement. If a mixer's references sound like one flat vocal with some reverb, they haven't done this work.
  • Afrobeats groove handling. Percussion layers, log drums, shaker patterns — these need width and separation without turning into wash. Listen for whether the percussion pushes the groove or just sits there.

Ask for two or three full songs, not a highlight reel. Reels are edited to show only the best eight seconds. Ask specifically: "Send me something you mixed that sounds like where I'm trying to go." If they can't produce it, that's not a scandal — it's information.

Ask about the process, not the plugins

The questions that actually predict a good experience are boring ones. Ask them anyway:

  1. What exactly do I get? Not "a mix." A stereo master-ready WAV at what sample rate and bit depth? Instrumental (TV) version? Acapella? Performance/clean edits? Stems by group? These matter later for sync, live shows, and remixes, and asking for them after the fact is how relationships sour.
  2. How many revisions, and what counts as one? A revision round means one consolidated list of notes, not eleven separate texts.
  3. What's your turnaround, and what happens if I'm late sending files? A real schedule has two sides.
  4. How long do you keep sessions and recalls? This is the recall question, and it's the one amateurs fail.
  5. Do you work to reference tracks, and how many? One or two references, chosen well, beat a folder of twelve.

On recall: a mixer who works in a properly labeled, fully in-the-box or well-documented session can reopen your song in six months — when a label, sync agent, or feature verse shows up — and make a targeted change in an hour. A mixer who can't will tell you the session is "gone" or quietly re-mix it from scratch and hand you something that doesn't match the version already on streaming. Ask directly how long sessions are archived and whether recall revisions after delivery are possible. Any honest answer beats a vague one.

How to send references (and what a good engineer does with them)

Bring one or two commercial songs, and be specific about what you want from each. "The vocal sits like this" is useful. "Make it sound like this" is not, because that record had a different singer, a different room, and different source material.

A capable mixer will do three things with your reference: level-match it against your rough so you're not fooled by loudness, name the two or three achievable qualities in it, and tell you plainly which parts are not reachable from your recordings. If the vocal was tracked into a laptop mic in a bedroom closet, no mix moves it to a $300-an-hour vocal booth. An engineer who tells you that before taking your money is the one worth hiring. If a mix engineer tells you your source needs a re-take, take the note seriously — that honesty is worth more than the session fee.

The myth worth killing: "A great mix engineer will fix it in the mix." Mixing is subtractive and corrective, not generative. It cannot un-clip a distorted vocal take, remove bleed baked into a single stereo bounce, retime a performance that was never in the pocket, or replace an arrangement that has four elements fighting for the same 200 Hz. What mixing does is take good decisions and make them undeniable. The most valuable thing an engineer does for you is often the sentence "re-cut that verse and I'll make it sing."

The red flags

These are the ones that show up over and over in bad hires:

  • "No revisions" or unlimited revisions. Both are wrong. Zero revisions means they're not planning to collaborate. Unlimited means they've never worked with an indecisive client and will resent you by round nine. A defined number — with a clear path to buy more — is the healthy answer.
  • No agreement of any kind. It doesn't need to be a twelve-page contract. It needs to be an email that states deliverables, revision rounds, turnaround, deposit terms, and credit. If someone resists writing it down, they're preserving the right to change the terms later.
  • Unclear deliverables. "I'll send you the mix" is not a deliverable. See the list above.
  • The one-size chain. If every song on their reel has the same vocal timbre, same reverb tail, and same brick-walled loudness, you're buying a template, not a mix. Your song will sound like their other songs.
  • Guaranteed results. "This will get you playlisted." "This will get you signed." Nobody controls that. Anyone selling it is selling you a feeling.
  • Loudness as the pitch. A mix that only impresses because it's louder than your rough is not a mix. Ask for a level-matched comparison, or match it yourself before judging.
  • Won't discuss what drives cost. Price varies enormously — number of tracks, condition of the source, tuning and editing needed, whether stems and alternates are required, deadline pressure. An engineer should be able to explain those factors clearly. What should worry you is a flat number quoted before hearing a single second of your music.
  • Ghosting during the free part. How someone communicates before you've paid is the best available preview of how they'll communicate after.

What working with Felix looks like

Felix mixes and produces out of Toronto, in person or fully remote, mostly R&B, hip-hop, and afrobeats — which means the low end, the vocal stacks, and the percussion are the parts he's spent the most time on, not an afterthought.

The process is deliberately unglamorous. You send the song and a reference or two. He listens first and tells you honestly what the mix can and can't do for this recording — including when a re-take will get you further than any amount of processing. You agree in writing on deliverables, revision rounds, and timing before anything starts. You get a defined set of revision rounds handled as consolidated note lists, not a text-message drip. Sessions stay archived and recallable, so when a feature, a sync placement, or a label version shows up months later, the record can be updated rather than rebuilt.

Cost depends on the song — track count, source condition, how much editing and tuning is in front of the mix, what alternate versions you need, and your deadline. That's a conversation, not a price list.

If you're deciding who to trust with a record you've spent months on, bring the rough and your references and book a call through the contact page. You'll get a straight read on what your song needs — and if that isn't Felix, he'll tell you that too.