Hire the Best Embedded Systems Engineers
in Georgia

Senior Embedded Firmware & Hardware Engineer with 10+ years of hands-on experience in microcontroller development. I specialize in full-cycle product development — from schematic design and PCB layout (KiCad) to production-ready firmware and post-launch support. My main platforms are STM32, ESP32, and nRF52. I work with bare-metal and FreeRTOS in C/C++, CMake/GCC build systems, and communication protocols (UART, SPI, I2C, USB, Ethernet/LwIP, BLE, IEEE 802.15.4). I also implement secure data transmission with mbedTLS. I take full ownership of projects: I cover technical risks, debug complex issues, and deliver reliable solutions ready for mass production. One of my biggest achievements is the Baxi Ampera Plus electric boiler — a mass-produced device sold in multiple countries. I refactored a large legacy firmware codebase (Keil → CMake/GCC + CI/CD), reduced graphics library by 4000+ lines, and significantly improved maintainability and reliability. Other notable projects: Wireless sensor networks for spacecraft transport containers (IEEE 802.15.4) Sensor nodes for UAV systems (STM32L4 + nRF52, BLE) HVAC thermal monitoring systems with BLE and TLS USB-HS CDC implementation on TI AM243x 1inch Hardware Wallet I prefer modern CMake-based builds and have experience with Zephyr RTOS. Always focused on low-power optimization, reliability, and bringing products from prototype to commercial success. Available for both short-term critical tasks and long-term development contracts. I’m committed to minimizing client risks and delivering working, production-ready solutions.

I am a hardware engineer with 10+ years of experience designing and bringing up complex embedded systems from technical specification and architecture to PCB design, validation, and production support. What I do: Design microcontroller- and microprocessor-based hardware for industrial, IoT, automotive-adjacent, and communication devices Own the full hardware cycle: component selection, schematics, PCB, bring-up, validation Work closely with embedded software teams on hardware–software integration Hardware & PCB Multilayer PCB design (up to 12 layers, controlled impedance, mixed-signal up to ~1 GHz) High-speed interfaces: USB 2.0 / 3.2, PCIe, Ethernet, SATA, HDMI, LPDDR4 DDR4 signal integrity modeling and timing analysis using Cadence Sigrity Power electronics: DC/DC, LDOs, power sequencing, thermal considerations EMC pre-compliance and debugging (including BCI testing) Embedded & Systems Platforms: STM32, ESP32, AVR, i.MX6, Raspberry Pi Firmware support: C, HAL, FreeRTOS, board bring-up, low-level debugging Embedded Linux systems and gateway-class devices Interfaces: SPI, I2C, UART, USB, SDIO, RS-232/485, Ethernet, PCIe Advanced & Domain Experience FPGA-assisted systems and hardware accelerator applications Automotive-grade hardware validation, DFMEA / FTA / FMEDA (Medini) Medical-related electronics (PPG, ECG, EMG) Industrial automation, IoT, telecom-related projects Tools & Bring-up Altium Designer, OrCAD / Allegro, SolidWorks, Git Professional lab equipment: oscilloscope, logic analyzer, signal generator Prototype assembly, rework, and debugging If you need an engineer who can design hardware, bring it up, debug real problems, and deliver a working prototype - I can help.

ATEH.tech is a engineering team specializing in Embedded Systems, AI Vision, custom electronics, and Linux-based devices. We specialize in projects that require expertise across electronics, embedded Linux, cameras, and system integration. We help founders, startups, and engineering teams turn ideas into working products — from architecture and hardware design to embedded software, prototyping, and manufacturing support. Over the last 20 years, we have developed custom electronics, embedded systems, carrier boards, AI cameras, and industrial devices for a wide range of industries including computer vision, industrial automation, telecommunications, security, and IoT. Our expertise includes: • Embedded Systems Development • NVIDIA Jetson (Orin NX, AGX Orin, Orin Nano, Xavier NX, AGX Xavier) • Rockchip Platforms (RV1126, RV1109 and others) • Custom Carrier Boards • AI Vision Systems • Sony Image Sensor Integration • Embedded Linux & BSP Development • Device Drivers & Camera Drivers • High-Speed PCB Design • FPGA-Based Solutions • Prototype Development • Manufacturing Support We typically work with founders, CTOs, and engineering teams who need an exexperienced partner capable of handling both hardware and embedded software development. Whether you are building an AI camera, industrial controller, smart sensor, edge AI device, or a completely new embedded product, we can help you reduce technical risks and accelerate develpment. 💡 AI Vision Focus (Core Strength) We specialize in building AI-driven devices: • Smart cameras and embedded vision systems • Custom carrier boards for NVIDIA Jetson (Orin / Xavier) • Image sensor integration (Sony IMX series, ISP pipelines) • High-speed embedded systems (MIPI, PCIe, DDR, Ethernet) • IoT, RF and Wireless devices (Wi-Fi, Bluetooth, LTE, LoRA) This allows me to solve complex problems where hardware, software, and AI intersect. 💡 What We Actually Solve A big part of our work is preventing expensive problems before they happen: • Unstable designs caused by poor power integrity • EMI/EMC failures during certification • High-speed signal issues (MIPI, DDR, PCIe) • Thermal and power bottlenecks • Prototypes that fail when scaling to production Fixing these later is expensive — I design systems correctly from the start. 🔧 What We Deliver (End-to-End) • System architecture and feasibility analysis • Schematic design and high-speed PCB layout (HDI, multi-layer) • Embedded software (MCU firmware, Linux BSP, drivers) • AI & vision hardware development (Jetson, ISP, sensors) • Hardware debugging and root-cause analysis • Prototype bring-up, validation, and optimization • DFM / DFT and manufacturing handoff (BOM, Gerbers, assembly docs) • Certification support (EMC, CE/FCC pre-compliance) ⚙️ Core Expertise • AI Vision Systems (smart cameras, edge AI, video pipelines) • Custom Carrier Boards (Jetson Orin, Xavier, Rockchip, NXP) • High-Speed Design (DDR, PCIe, MIPI CSI/DSI, USB 3.x, Ethernet) • RF & Wireless (Wi-Fi, BLE, LTE, GNSS) • Power Electronics (DC/DC, battery systems, PoE) • FPGA & SoC-based systems 🧪 Prototyping & Lab We maintain a fully equipped engineering lab: • Oscilloscopes, logic analyzers, spectrum analysis • PCB prototyping, soldering, rework • CNC and 3D printing • In-house bring-up and validation This allows me to identify and fix issues early — before they reach production. 🧩 How We Work • Clear requirements and system-level thinking • Structured development with milestones • Regular updates and transparent communication • Focus on reliability, manufacturability, and cost efficiency No guesswork. No disappearing after the prototype. 📦 What You Get • Production-ready hardware — not just a prototype • Stable, predictable system behavior • Reduced risk of redesign and manufacturing issues • Faster time-to-market • Complete documentation for scaling production 🏁 Let’s Build Something Serious If you're building a complex AI device, smart camera, or embedded system — or fixing hardware that doesn’t behave in production — We can help. Send me a message with your idea or current challenges — I’ll respond with the right questions and a clear next step.

Feedback: 💭 "Roman has a vast knowledge of what he is doing and really good at what should be done" 💭 "He has good social skills; is patient and a good communicator. His skills are very much in demand" 💭 "If some solution does not work or is not possible Roman will find another way to get the task done" 💭 "His research is solid, resulting in a good overview of possibilities to (re-)direct the project" Introduction: I'm a software engineer with 20 years of experience and more than a decade in architect and tech lead roles. My strength lies in combining deep technical expertise with a big-picture understanding of how modern software systems evolve over time. I help businesses build reliable, scalable, and maintainable systems -- whether by designing new solutions from the ground up or by rethinking and refactoring existing systems into clean, resilient architectures. Over the years, I have seen how software can become either a long-term business liability or a strong foundation for growth, and my goal is always to help clients achieve the latter. My experience is backed by a strong track record of completed projects, enterprise-level engineering work, and positive customer feedback. After building a successful career in enterprise software, I now work as a freelancer with clients around the world, helping them solve complex technical challenges with clarity, reliability, and long-term thinking. Expertise: Streaming & Multimedia: 🤖 Live bots and agents (stt, tts, livekit, azure foundry) 📠 IP Telephony (asterisk, freeswitch, twilio, sip, voip) 📹 Streaming protocols (webrtc, hls, dash, rtmp, rtsp) 📺 Media applications (ffmpeg, gstreamer, opencv) 🎛 Media services (kurento, mediasoup, janus, wowza) 🧩 Codecs and Muxers (h.264/avc, h.265/hevc, av1, vp9) 📐️ GPU programming (opengl, opencl, cuda, vulkan, hip) DevOps & Cloud & IoT: ☁️ Cloud Infrastructure (aws, ansible, terraform, kubernetes) ⛓ Build and test automation (jenkins, gitlab, ci/cd pipelines) 🔗 Tunneling protocols (openvpn, wireguard, gre, ipsec) 🧶 Bulk networking (dpdk, bpf/ebpf, xdp) 📦 Virtualization (kvm, qemu, crosvm, virtio) 🔋️ IoT / Embedded (raspberry pi, jetson, nordic, ble) 💻️ Linux deployment (containers, docker, yocto) Software development: 🐍️ Python (flask, asyncio, fastapi, websockets) 🕸️ Node.js (javascript, typescript, npm, react) ☕ Java (maven, hibernate, spring, springboot) 🍀️ C/C++ (qt, cmake, meson, kernel, ndk, maths) 🦀 Rust (cargo, actix, tokio, rayon, solana) 🛠 Reverse engineering (disassembling, protocols)

I help companies build performance-critical software for a wide range of platforms, including backend, embedded, desktop, and mobile. My core expertise is turning complex low-level requirements into reliable and scalable architectures and implementations. A significant part of my work has focused on low-latency audio systems and streaming software. Project Highlights: 1. Roc Streaming — co-founded and maintain an open-source real-time streaming ecosystem, including embedded components, REST services, and mobile apps. 2. ViveSound — co-founded ViveSound and built a commercial desktop audio engine, plus desktop and mobile applications. 3. Online jamming platform — built embedded Linux software for a real-time music collaboration platform with low-latency audio streaming. 4. Airborne synchronization — built embedded Linux software for acoustic-signature-based playback synchronization. 5. Distributed playback sync — built distributed multi-receiver playback synchronization over high-jitter Internet connections. 6. Trading data proxy — built a high-load concurrent HTTP proxy in Go for trading data, including stream routing, balancing, and caching.

Machine Learning and AI Engineer. 8+ years shipping applied AI into production: LLM agents and RAG, NLP, speech, computer vision, and CUDA. I build systems that hold up in production, not demos, and every claim below is a number I measured. LLM and NLP - Semantic CV-to-job matching for an EU job board: 10,001 CVs against 10,003 vacancies, 100M cosine pairs in about 18s on one RTX 4090 (bge-m3), then a two-stage LLM re-rank that spreads the top 10 from 0.93 down to 0.71 where the raw cosine moves only 0.03. - LLM CV builder: messy PDF / DOCX / LinkedIn text into a strict Pydantic record bound to the client's 674-value taxonomy, so invalid output is structurally impossible. Emits a recruiter-ready LaTeX PDF and database-ready JSON. - Crypto signals from tweet embeddings (live client program): Qwen3-Embedding-8B on an A40, 1M+ tweet vectors in Qdrant, 44 symbols scored every 60 seconds, 4.65M events ingested, 10-day zero-error run. Speech - On-prem call-centre speech-to-text, in production 2020-2022, feeding fraud detection and call routing. Benchmarked on hand-labeled calls: Whisper 20% median WER, Vosk 28%, Google 32%. - Diarized transcription toolkit, TTS, and cloned-voice agents. Computer vision and CUDA - Multi-camera 3D capture for film relighting: 33 datasets delivered, 3-32 synchronized cameras each, GPU space carving at 1.98 s/frame versus 65 s in vectorized NumPy (about 33x). 5.0 Upwork review. - GPU brute-force passphrase recovery: about 58,000 candidates/s on one RTX 4090, 160x over CPU, sharded across multiple GPUs and machines. - Open-set retail robot vision on ROS 2; banknote detection at the cash desk. YOLO, Grounding DINO, RealSense. How I work Anti-hallucination is enforced in the code path, not in a prompt: strict schema validation, retry on violation, closed enum vocabularies. Success criteria are pre-registered before results exist, models are judged on time-ordered splits, and negative results get reported rather than buried. Stack - LLM and agents: OpenAI API, Claude Agent SDK + MCP, LangChain / LangGraph, RAG, vLLM, Hugging Face Transformers, fine-tuning, structured outputs with Pydantic + instructor, hand-written tool-use loops, chatbots on Telegram and web - NLP and search: embeddings, semantic search, vector databases (Qdrant), clustering, topic discovery, classification, multilingual pipelines - Speech: Whisper, Vosk / Kaldi, Deepgram, Google Cloud STT, TTS, speaker diarization - Vision: OpenCV, YOLO, Grounding DINO, TensorFlow Object Detection, RealSense RGB-D, ROS 2, NVIDIA Jetson - Classic ML: PyTorch, TensorFlow, scikit-learn, CatBoost, forecasting, time-series validation, reinforcement learning - GPU: CUDA C++, PyCUDA, Nsight Compute, multi-GPU sharding - Data and infra: Python, FastAPI, Docker Compose, PostgreSQL, Redis, BigQuery, pandas / NumPy / Parquet, n8n, Linux VPS, GCP Full case studies for each project above, with architecture diagrams and honest results including what did not work, are in my Portfolio section below. Tell me what you are building and I will tell you plainly whether I am the right fit for it.