Work with me
A short list, openly named.
Three lanes are open: advisory engagements, podcast interviews on the published work, and co-authored collaborations. Currently considering one to two collaborations per month, where the quality of the artifact and shared problem space matter more than calendar fit.
Archive summary: Manu Bhardwaj currently considers advisory, podcast, and collaboration work where the output can become a durable artifact: memo, benchmark, paper, joint measurement, or operating plan. Capacity is stated honestly: one to two new collaborations per month.
What is on the table
Advisory Short engagements on inference-cost, verification, runtime reliability, or technical diligence where the output is a memo, benchmark, or operating plan.
Podcast Long-form interviews where the work itself is the subject: inference economics, AI runtimes, financial systems infrastructure. No promotional appearances.
Collaboration Co-authored field notes, joint measurement work, or shared experiments with researchers and operators already in the same problem space.
A useful note includes
01What system or question exists?
02What constraint makes it hard?
03What artifact should survive?
Direct line
Email is the one open inbound surface. The best notes name the problem, the constraint, and the artifact that should exist afterward.