The company is part of the Iliad group and builds and runs its own 5G network, from O-RAN radio units and distributed units to the user plane and the Cloud RAN environment.
The DU physical layer and 5G fronthaul work under strict real-time limits: each 0.5 ms slot arrives from the radio unit over eCPRI and must be processed on GPU, DSP or SIMD CPU and passed to the upper layers in time.
The role combines 5G signal processing, embedded software and high-performance networking, and you work with real IQ data, radio units and production limits, not only simulations.
AI-assisted algorithms are one tool among others. The team adopts them only when they measurably beat classical methods on BLER, throughput or latency.
Recruitment steps: a 15-minute phone screen with recruiter Steve, a 30-minute interview with hiring manager Vincent, a 60-minute technical interview, and a 60-minute HR interview with Sarah and Steve.
Responsibilities
Design and optimise physical-layer and fronthaul processing for an O-RAN 7.2x system, from the algorithm and simulation stage through real-time implementation to integration with radio units.
Build the 5G uplink receiver: channel estimation, interpolation, MMSE equalisation, MIMO detection, LLR computation and LDPC decoding.
Implement PRACH detection, timing estimation and massive MIMO beamforming.
Handle O-RAN 7.2x C-plane and U-plane traffic, including eCPRI timing windows and IQ compression.
Write real-time L1 functions in C on Linux, and speed up heavy kernels with CUDA, DSP or SIMD.
Build the eCPRI datapath on DPDK using poll-mode I/O, rteflow, dedicated queues, zero-copy buffers and hardware timestamps.
Tune the full processing chain for latency, memory, cache behaviour and compute load within the 0.5 ms slot budget.
Test algorithms in link-level simulations with 3GPP TDL/CDL channel models, then with real IQ captures and O-RAN radio units.
Measure BLER, throughput, latency and CPU, GPU or DSP load to compare classical and AI-assisted approaches.
Explore neural channel estimation and neural receivers where they could beat an established baseline.
Create regression tests and CI that catch drops in performance or accuracy.
Help decide architecture, including GPU versus DSP or SIMD, hardware selection and accelerator integration.
Work closely with the radio unit, RAN, core network and operations teams.
Over time, take full ownership of technical topics, from the mathematical model and implementation choices through to validation on the DU test bench.
Review code, write design documents, and contribute to open-source projects such as DPDK, Linux and U-Boot.
Requirements
Experienced engineers and high-potential graduates are both considered, provided they have solid foundations in mathematics, signal processing or embedded software.
A degree in telecommunications, applied mathematics, embedded systems, computer science or a related field at engineering, Master of Science or PhD level.
Solid grounding in linear algebra, probability and digital signal processing.
Good working knowledge of OFDM, MIMO, Fourier transforms, channel estimation, detection or channel coding.
C programming on Linux, with an interest in memory layout, multithreading, cache behaviour and performance.
Python for simulation, testing and data analysis, using NumPy and SciPy.
Interest in at least one of: 5G/LTE physical layer, O-RAN fronthaul, real-time embedded software, GPU programming, DSP/SIMD optimisation or AI for signal processing.
Ability to turn maths into code, define meaningful metrics and check results experimentally.
Good written English for specifications, design documents and code review. French is a plus.
Experience with CUDA, cuBLAS, cuFFT, AVX2, AVX-512, Arm NEON/SVE or fixed-point arithmetic is valuable, though mastering all of them is not required.
Knowledge of DPDK, eCPRI, PTP, O-RAN 7.2x, 3GPP channel models or massive MIMO is a plus.
Most important is your ability to work through hard technical problems, build solutions you can measure, and explain the reasoning behind your choices.
Conditions
Up to two days per week can be worked remotely, with one day being the minimum.
Secure bicycle parking is available on site.
The company's internal mobility options across the Iliad group are open to you.
The company is committed to non-discrimination and to preventing harassment, and welcomes applications regardless of background, age, origin, gender or sexual orientation.
Benefits
Profit-sharing and incentive scheme linked to group performance.
Your home internet plan with the company is reimbursed.
Up to 70% of public transport costs is paid back to you.
Meal vouchers from Swile worth €10 per working day, with the company covering 60%.
Regular training, with access to the company's internal training school.
Comprehensive health and income protection cover.
Time Savings Account, known in France as a Compte Épargne Temps or CET.
Employee benefits through the CSE, including gift vouchers, discounts and exclusive offers.
On-site restaurants and cafeteria.
Work in a distinctive office in a landmark building that is part of Paris's architectural heritage.