Current Position & Biography
Dakai Kang is a research scientist at Sei Labs, according to his personal biography, which dates the appointment to June 2026. He describes recently completing his computer-science doctorate at UC Davis and working on fault-tolerant consensus, transaction ordering and parallel execution. He also identifies himself as an architect of ResilientDB. The United States field reflects Sei Labs’ advertised operating locations in New York City and San Francisco; it does not assert Kang’s personal worksite, residence or nationality. His earlier UC Davis affiliation is not treated as his current employment.
Notable Contributions
Kang coauthored FairDAG: Consensus Fairness over Multi-Proposer Causal Design with Junchao Chen, Tien Tuan Anh Dinh and Mohammad Sadoghi. The 2025 PVLDB paper studies transaction-order fairness in consensus systems with several block proposers. FairDAG places fairness mechanisms over directed-acyclic-graph consensus, seeking to reduce the influence of a single proposer while preserving throughput. Its two evaluated variants address different ordering guarantees. The authors compare fairness quality and performance under their stated models and experiments; this is not proof that all transaction-order manipulation or extractable value disappears in a deployed financial application.
Why Katharos.ai Is Following Them
Katharos.ai is following the distinction between agreeing on an order and agreeing on a fair order. A ledger may remain internally consistent while its ordering rules advantage some participants over others. Research that makes the fairness criterion explicit helps readers ask more precise questions about decentralized-market infrastructure. The important issue is not simply how many transactions a system can process, but which choices the protocol permits before those transactions become final.
Areas to Follow
Areas to follow include independently reproducible evaluations, treatment of network delays, implementation assumptions and how ordering mechanisms interact with execution. Useful coverage should distinguish a mathematical fairness property from a broader claim about fair market outcomes. These are editorial priorities rather than assurances about token performance, censorship resistance in every circumstance or protection against every trading strategy.
Reporting Scope
The current appointment and project role are attributed to Kang’s personal website. Company-level geography follows an employer-published job listing, not a claim that Kang holds the advertised product-management job. The technical discussion credits collaborative research and does not assert that every paper mechanism is deployed on Sei. Neither the university affiliation printed on the paper nor the academic clothing in the photograph establishes current employment.
Sources for Further Reading
https://dakaikang.github.io/ https://www.vldb.org/pvldb/vol19/p265-kang.pdf https://jobs.ashbyhq.com/sei-labs/0d0886c2-cf9e-485e-90bf-2b77282fd45e
Photo Credit
Portrait via Dakai Kang’s personal website; photographer and capture date unidentified. Full image reduced with neutral padding; original metadata retained. Limited factual editorial identification, not endorsement; no open license asserted. The source page carries a generic copyright notice in Kang’s name; public availability is not unrestricted permission. The modest complete-image presentation identifies the researcher without altering his face. The original JPEG, including EXIF, Photoshop, ICC color-profile and JFIF metadata, is retained and those metadata segments remain attached to the derivative.
Editorial Disclaimer
This profile summarizes information from the cited sources. Katharos.ai does not endorse the subject or guarantee every source claim’s authenticity.