Location: Frankfurt am Main, Germany (On-site) Type: Full-time (initially 3 years)
Natural Intelligence (NISYS GmbH), a newly founded deep-tech startup at the intersection of neuroscience and machine learning, invites applications for a Lead Hardware Engineer to build the first physical implementations of our brain-inspired neural networks.
Join an ambitious team that turns new neuroscientific insights into reliable, efficient next-gen AI systems. Our networks compute with coupled oscillators, and you will take our hardware story to the next level. The company has successfully raised a multi million Euro seed round and is fully funded for the coming years.
The Role
As a foundational member of our engineering team, you own the path from a model trained on a digital platform to a working device on the bench. Our Research Scientists provide you with the models and our world-class advisors in analog chip design experts support you on circuit and silicon questions. You are responsible for the architecture, the integration, the schedule, and the result, and you turn measurements into constraints that informs the scientists in their future modeling efforts. You will help decide what the hardware roadmap looks like and as our hardware effort grows, you will hire and lead the engineers who join the hardware team.
Responsibilities
- FPGA Prototyping: Design, implement, and bring up oscillatory neural networks on FPGA, from fixed-point modeling of the oscillator dynamics to real-time inference on live sensor input. You choose the platform, write the RTL, close timing, and make it work
- Analog and Mixed-Signal Path: Drive the transition from FPGA to analog hardware. Build board-level mixed-signal prototypes. Together with our analog chip designers and external partners, define block-level specifications for an analog chip and plan test chips. You own the system specification, the analog designers own the circuits
- Hardware-Model Co-Design: Work with the research scientists to make models hardware-ready. Quantify what precision, dynamic range, noise, and mismatch the network tolerates, and communicate those limits into model design and training
- Bring-Up, Test, and Characterization: Own the lab. Bring up boards and chips, characterize them, and compare measurements against a digital twin
- Technical Interface: Be the single point of contact between scientists, analog chip designers, and infrastructure engineers. Turn scientific questions into engineering tasks and engineering results into answers scientists can use
- Project Lead: Set the hardware roadmap, define milestones, and report progress and risks to the Chief Scientist. Represent the hardware effort toward academic and industry partners
- Team Building: Define the roles the hardware team needs next, interview candidates, and onboard and lead the engineers who join
Your Profile
We are looking for an engineer who has built hardware that worked, on both sides of the analog-digital boundary, and who wants to lead by doing.
Essential Technical Requirements
- Education: M.Sc. or Ph.D. in Electrical Engineering or a related field (e.g. Physics, Computer Engineering)
- Experience: 5+ years of hands-on hardware engineering, with a track record of designs that reached working prototypes or products
- FPGA Expertise: Multi-year experience with FPGA design in VHDL or SystemVerilog, including DSP-style datapaths, fixed-point arithmetic, timing closure, and bring-up on real boards. Fluent with at least one major toolchain (AMD/Xilinx Vivado, Intel Quartus). High-level synthesis experience is a plus
- Analog and Mixed-Signal Design: Working knowledge of analog circuit design (amplifiers, integrators, oscillators, filters, transconductance stages) and mixed-signal system design (ADC and DAC interfaces, signal integrity). You have simulated circuits in SPICE and built and measured them. Tape-out experience is a strong plus, not a requirement
- Lab Skills: Comfortable with oscilloscopes, signal generators, spectrum analyzers, and debugging across board, firmware, and software
- Software: Strong Python for modeling, simulation, and test automation. You can read PyTorch or JAX model code and turn it into a hardware specification
- Leadership: Experience leading a small engineering team or a hardware project through to a working demonstrator
Bonus
- Experience with neuromorphic, analog, or other physics-based computing
- Background in dynamical systems, recurrent neural networks, or signal processing
- Experience with CMOS IC design flows, multi-project-wafer runs, or working with a foundry
- Experience with embedded systems or real-time signal processing on audio or sensor data
Soft Skills
- Ownership: You take a vague goal, turn it into a plan, and deliver working hardware without waiting for someone to hand you a specification
- Translation: You can explain quantization to a neuroscientist and metastability to a circuit designer, and you enjoy being the person who does both
- Judgment Under Uncertainty: You know when a prototype is good enough to learn from and when to stop and rebuild
- Communication: Excellent written and verbal communication in English. German is a plus, but not required
- Collaborative Mindset: You enjoy an interdisciplinary environment where physics, biology, and computer science meet
What We Offer
- Impact: You build the first physical instances of a new computing architecture. Your design decisions define the hardware line of the company
- Environment: A creative setting combining academic rigor with entrepreneurial agility. You will work within a network of renowned research institutes while enjoying the fast-paced execution of a startup
- Autonomy: A direct line to the Chief Scientist, a short decision path, and control over how the hardware effort works
- Resources: A dedicated hardware lab, access to modern compute clusters, and an infrastructure team to support your work
- Growth: Lead the hardware team as it grows. Collaborate with international partners and present your work at conferences
- Package: Competitive compensation plus benefits
About Natural Intelligence
Natural Intelligence (NISYS GmbH) is a research-driven startup founded by Dr. Felix Effenberger and Prof. Wolf Singer. Our mission is to use the principles of brain dynamics, self-organization, and oscillatory computation to create a new generation of intelligent systems. Based in Frankfurt am Main, Germany, we collaborate closely with local academic institutions, including the Ernst Strüngmann Institute.
How to Apply
The review of applications will begin immediately and continue until the position is filled.
Please send your application to: [email protected]
Please submit:
- Your CV (referees on request)
- A brief cover letter describing one piece of hardware you built that worked, what went wrong on the way, and what you did about it
- Optional: a link to your GitHub / portfolio or a sample of your RTL or circuit work, if you are free to share it