Muriel Medard

Muriel Médard (born February 1, 1968) is a co-founder of Optimum and a French-American information theorist and electrical engineer whose research has helped develop network coding and reliable communication systems.^[2] She is the NEC Chair of Software Science and Engineering in the Massachusetts Institute of Technology’s (MIT) Department of Electrical Engineering and Computer Science (EECS), where she leads the Network Coding and Reliable Communications Group.^{[1] [2]}​^[3]​

Early Life and Education

Médard attended Lycée International de Saint-Germain-en-Laye from 1978 to 1985 and also took home classes in Russian. She studied at MIT from 1985 to 1995, earning bachelor’s degrees in electrical engineering and computer science and in mathematics in 1989, followed by a bachelor’s degree in humanities and a Master of Science in 1991. She completed an Sc.D. in electrical engineering and computer science in 1995; her doctoral adviser was Robert G. Gallager.^[1][3][4]

Career

Academic Career

Following the completion of her doctorate, Muriel Médard joined MIT Lincoln Laboratory in 1995. She initially worked as a postdoctoral researcher and later as a member of the technical staff, remaining at the institution until 1998. In August of that year, she became an assistant professor at the University of Illinois at Urbana-Champaign, where she worked until December 1999, which served as a foundation for his later work with cryptocurrencies.

In January 2000, Médard joined the Massachusetts Institute of Technology (MIT) as an assistant professor. She was promoted to associate professor in 2003 and professor in 2005. Between 2014 and 2022, she held the Cecil H. Green Professorship in Electrical Engineering and Computer Science (EECS). In July 2022, she assumed the NEC Professorship of Software Science and Engineering at MIT, where she continues to hold a faculty position.

At MIT, Médard directs the Network Coding and Reliable Communications Group within the Research Laboratory of Electronics. The group conducts research in network coding, information theory, wireless communications and optical networking. Research from the group has also informed the development of data propagation methods used in decentralized systems, including those associated with Optimum.

Network Coding and Random Linear Network Coding

Médard's research includes the development and application of network coding, a communication method in which network nodes combine information from received packets before transmitting it to other nodes. Random linear network coding (RLNC) extends this approach by allowing nodes to independently generate and transmit random linear combinations of available packets. This eliminates the requirement for each node to coordinate with a central authority or decode complete packets before forwarding information.

In 2003, Médard and her colleagues introduced RLNC at MIT. Her publications in this area include An Algebraic Approach to Network Coding (2003), co-authored with R. Koetter, which presented a linear-algebraic framework for representing network flows. In 2006, she co-authored A Random Linear Network Coding Approach to Multicast with T. Ho, R. Koetter and other researchers. These studies addressed the mathematical formulation and implementation of coding techniques for data transmission across network nodes.

Her research also examined the use of network coding in wireless communication systems. In 2008, she co-authored XORs in the Air: Practical Wireless Network Coding with Sachin Rajsekhar Katti, Hariharan Shankar Rahul, Wenjun Hu and other researchers. The publication examined how wireless networks could use packet combinations when multiple nodes receive the same transmission, rather than relying exclusively on retransmission of individual packets.

Reliable Communication and Network Operation

Médard's work on communication reliability examined the use of coding techniques in packet networks subject to transmission losses. In On Coding for Reliable Communication over Packet Networks, presented in 2004 with Desmond S. Lun and Michelle Effros, the authors investigated a method in which intermediate nodes transmit random linear combinations of received packets without first decoding an entire block. The approach addresses both unicast and multicast communication over lossy networks. Its coding and decoding procedures have polynomial computational complexity.

The study also considered links with average-rate packet arrivals and correlated losses. Its mathematical analysis related the transmission of innovative information to the movement of jobs through a queueing network. An extension published in 2005 expanded the arrival model and derived error exponents for a Poisson traffic model with independent losses.

In Minimum-Cost Multicast over Coded Packet Networks, Médard and her collaborators examined the cost of multicast transmission in networks where outgoing packets can be generated from previously received information. For static multicast groups, they formulated the problem as an optimization task solvable in polynomial time and presented decentralized algorithms that could operate alongside network-code construction. For multicast groups with changing membership, they used dynamic programming to formulate the problem. The study considered both wireline and wireless communication systems.

Other publications examined the integration of network coding into transport protocols and packet queue management. Network Coding Meets TCP investigated changes to the Transmission Control Protocol (TCP) that would allow sources to transmit random linear combinations of packets within their congestion windows. Receivers could acknowledge individual degrees of freedom without necessarily reconstructing an original packet immediately. The authors reported higher throughput than conventional TCP over lossy wireless links and examined the protocol's fairness and soundness properties.

In ARQ for Network Coding, Médard and her co-authors presented a "drop-when-seen" algorithm for packet queue management. The method synchronizes the sender's physical queue with the number of degrees of freedom still required by the receiver. The study compared its expected queue size with that of a "drop-when-decoded" approach.

Her research has also covered wireless channel capacity and communication under different transmission conditions. A publication from 2000 examined the effects of perfect and imperfect channel knowledge on capacity in wireless communications. In collaboration with Daniel E. Lucani and Milica Stojanovic, Médard developed an approximate model of underwater acoustic communication. The model examined the relationships between transmission power, frequency band, bandwidth, distance and capacity, while accounting for propagation loss and ambient noise.

The same collaboration investigated RLNC in time-division-duplex packet-erasure channels, where nodes alternate between transmission and reception rather than performing both simultaneously. The proposed method examined the number of coded packets that could be transmitted between listening periods. The researchers also analyzed the expected time required to complete transmissions under this arrangement.

Security, Privacy and Decoding

Médard's research includes mathematical approaches to security and privacy in information networks. In 2007, she collaborated with Luísa Lima and João Barros on a study that developed an algebraic criterion for evaluating network-coding security in systems where protocol-compliant nodes could act as "nice but curious" eavesdroppers. The researchers examined the relationship between network topology and information security, including configurations involving complete directed acyclic graphs.

In 2014, Médard, Ali Makhdoumi and Salman Salamatian investigated the relationship between privacy and data utility when disclosed information is correlated with private user data. Their research formulated the log-loss problem as a Privacy Funnel optimization task and introduced a locally optimal greedy algorithm. The method was evaluated using data from the United States census.

In 2020, Médard collaborated with Amit Solomon and Ken R. Duffy on SGRAND, a soft-information maximum-likelihood decoding algorithm developed from the hard-detection GRAND decoder. The algorithm was designed for high-rate, short-length block codes. Its evaluation included cyclic redundancy check (CRC)-aided Polar codes used in 5G New Radio control-channel communications. The researchers compared its performance with CRC-Aided Successive Cancellation List decoding.

Professional Leadership and Academic Service

Médard served as president of the IEEE Information Theory Society from 2012 to 2013. Her other responsibilities within the organization have included membership on its board of governors, where she served for eleven years. She has also co-chaired technical program committees for the International Symposium on Information Theory, CoNEXT, WiOpt and WCNC.

Her professional service within IEEE includes chairing the organization's Medals committee and serving as the inaugural chair of the Millie Dresselhaus Medal.

Médard has held editorial positions in several academic publications. She served as editor-in-chief of the IEEE Journal on Selected Areas in Communications and has worked as an editor or guest editor for other IEEE journals. Her positions have also included editor-in-chief of IEEE Transactions on Information Theory and membership on the inaugural steering committees of IEEE Transactions on Network Science and IEEE Journal on Selected Areas in Information Theory.

She established Women in the Information Theory Society (WithITS) and the Information Theory Society Mentoring Program. In 2017, she received the Aaron Wyner Distinguished Service Award for her activities in the society. In 2013, she received MIT's inaugural EECS Graduate Student Association Mentor Award, selected through a student vote.

Her academic supervision includes more than 40 master's students, over 20 doctoral students and more than 25 postdoctoral fellows. From 2002 to 2007, she also served as undergraduate Faculty in Residence in two MIT dormitories. Her institutional responsibilities included several faculty committee positions, among them chair of the Institute Committee on Student Life.

Between 2003 and 2006, Médard was a member of the Defense Advanced Research Projects Agency's (DARPA) Information Science and Technology (ISAT) advisory group. Since 2015, she has served as treasurer of the International School of Boston's Board of Trustees.

Industry Experience and Consulting

Alongside her academic career, Médard has held consulting, scientific leadership and advisory positions in technology companies. She co-founded CodeOn, also identified in her professional experience as Code On Technologies, and has served as a consultant for the company since January 2010. She is also identified as a co-founder of Steinwurf, where she has held consulting and scientific positions.

Between 1999 and 2001, she served as an advisor to Sandial Systems. From 2000 to 2002, she worked as a consultant for Sycamore Networks and Vanu, Inc. From July 2008 to August 2009, she held the position of chief scientist at Blackwave, formerly known as Acinion.

Médard worked as a consultant for Steinwurf from January 2010 to July 2015. She later returned to the company as chief scientist in January 2021, holding the position until June 2024. From June 2024 onward, she has continued working with Steinwurf as a consultant.

Her other industry positions include chief scientist at NanoPing from 2021 to June 2024, followed by a consulting role at the company. She joined Nokia Bell Labs' advisory council in January 2022 and has served as a member of the council. From October 2023 to September 2024, she was also a member of Ddosify's technical advisory board.

Optimum

In February 2024, Médard co-founded Optimum alongside Kishori Konwar and Kent Lin. She serves as the company's chief executive officer. Optimum develops memory infrastructure for blockchain networks using random linear network coding (RLNC). Its work addresses data propagation, network speed, robustness and throughput in decentralized applications.

The company's technology draws on research conducted by Médard's group at MIT, particularly in network coding and information transmission. Its systems apply these methods to decentralized networks, where latency and bandwidth influence the distribution of information among network participants.

In April 2025, Optimum announced an $11 million seed funding round led by 1kx. Other participants included Robot Ventures, Finality Capital, Spartan, CMT Digital, SNZ, Triton Capital, Big Brain, CMS, LongHash, NGC, Animoca, GSR, Caladan and Reforge, among other investors.

The company also reported the launch of a private testnet involving 40 validator partners and connections to more than $24 billion in staked ETH. It later announced the launch of mump2p, its first data-propagation protocol, on Ethereum mainnet. At the time of the announcement, the deployment involved more than 20 validators and over 1.4 million staked ETH.

Awards and Recognition

Médard was elected an IEEE Fellow in 2008 for her contributions to wideband wireless fading channels and network coding. In 2018, she became a Fellow of the U.S. National Academy of Inventors. She was elected to the U.S. National Academy of Engineering in 2020 and became a Fellow of the American Academy of Arts and Sciences in 2021. In 2022, she was elected to Germany's National Academy of Sciences Leopoldina, followed by her election to the Royal Academy of Engineering in 2025.

Her academic distinctions also include honorary doctorates from the Technical University of Munich, Aalborg University and the Budapest University of Technology and Economics.

Her individual awards include the IEEE Vehicular Technology Society's James Evans Avant Garde Award in 2016, the IEEE Communications Society's Edwin Howard Armstrong Achievement Award in 2017 and the IEEE Koji Kobayashi Computers and Communications Award in 2022. In 2026, she received the IEEE Richard W. Hamming Medal for contributions to coding for reliable communications and networking.

Several research publications co-authored by Médard have also received academic awards. Research related to random linear network coding received the IEEE Communications Society and Information Theory Society Joint Paper Award and the William R. Bennett Prize in Communications Networking in 2009. Related research published during the 2000s received the ACM SIGCOMM Test of Time Paper Award in 2018.

Other paper awards include the IEEE Leon K. Kirchmayer Prize Paper Award in 2002 and a 2019 Best Paper Award from IEEE Transactions on Network Science and Engineering. According to her MIT profile, most of her prize-winning publications were co-authored with students from her research group. ^{[1] [2] [3] [4] [5] [6]}

Interviews

Decentralized Coding and Blockchain Data Propagation #01

In an interview published on the Bitcoin.com - Official Channel YouTube channel on December 6, 2025, MIT professor Muriel Médard discussed Optimum, a project developing a decentralized memory layer for blockchain networks. She described how the system uses random linear network coding (RLNC) to distribute data among nodes by transmitting combinations of information rather than repeatedly sending identical data packets.

According to Médard, Optimum functions as an additional infrastructure layer that can operate alongside existing blockchain protocols without requiring changes to their execution environments. She reported that tests on the Ethereum Hoodi testnet achieved data propagation speeds approximately six times higher than those of existing approaches. She also cited private test environments in which the system processed data at rates 10 to 20 times higher than conventional implementations.

Médard stated that the coding mechanism also supports data verification and resistance to Byzantine faults, which can occur when network participants provide incorrect or conflicting information. The project initially targeted Ethereum validators and block builders, with its architecture allowing participation by additional nodes. She described Optimum as blockchain-neutral, with the intention of supporting different networks through a modular integration approach. ^[7]

Decentralized Network Models #02

In a TEDxMIT presentation titled Three Networks That Changed the World, published on the TEDx Talks YouTube channel on July 13, 2026, Muriel Médard examines the development of network abstractions, using electrical grids, telecommunications, and the internet to explain the principles underlying decentralized computation.

Médard explains how electrical grids provide access to power through standardized sockets, allowing electricity to be consumed without requiring users to know its generation source or the infrastructure responsible for its distribution. She draws a parallel with telecommunications, describing the transition from direct telephone connections to interconnected networks operated through switching systems. She also discusses the introduction of TCP sockets, which standardized communication between devices over the internet and separated network access from individual physical connections.

The presentation extends this comparison to computing services. Médard observes that conventional cloud computing requires users to access resources managed by specific providers, such as Amazon and Google, which maintain control over the infrastructure and associated data. She presents Web3 and blockchain networks as an approach to organizing computation in which data and computing resources are associated with users rather than exclusively with service providers. She outlines four layers in this architecture: application, execution, settlement, and consensus.

According to Médard, decentralized computation still faces challenges in data access and distribution. Although virtual machines, including the Ethereum Virtual Machine (EVM), provide environments for executing decentralized applications, she identifies the underlying network infrastructure as an area requiring further development. Her research in decentralized coding, a field within information theory, addresses methods for representing and distributing data with considerations for efficiency, reliability, and security. She relates this research to the development of network systems in which data access and computation are not restricted to individual service providers. ^[8]