Software Engineer (AI)

Fresher

Software Engineer (AI)

Job Description: We are looking for a curious and driven AI Research Developer to join our expanding team. In this role, you will explore prototype and benchmark emerging AI techniques while working with our engineering teams to turn promising research into deployable software. This position suits candidates who enjoy reading papers, reproducing results, and translating them into working code for real products across cloud and edge environments. Key Responsibilities: Track recent developments in generative AI, agentic systems, and model optimization, and assess their relevance to ongoing and upcoming projects. Reproduce results from research papers and open-source repositories and build proof-of-concept prototypes. Design and run experiments: dataset preparation, training/fine-tuning runs, ablation studies, and evaluation against defined metrics. Benchmark models for accuracy, latency, memory footprint, and throughput on target hardware. Work with product and engineering teams to hand over validated approaches, including documentation, reference implementations, and integration support. Write clear technical reports, internal documentation, and presentations for both technical and non-technical audiences. May have to interact with customers to understand problem statements and propose feasible AI approaches. Participate in code reviews, experiment reviews, and team knowledge-sharing sessions.

Skills and Qualifications Must Have: Bachelor's or Master's degree in Artificial Intelligence and Machine Learning, Computer Science, Electronics, Mathematics, or a relevant field. Strong programming skills in Python. Solid grounding in machine learning and deep learning fundamentals: activation functions (ReLU, GELU, SiLU, Softmax), loss functions, optimizers, regularization, and evaluation metrics. Good understanding of Transformer internals (Attention, MLP, LayerNorm, RoPE) and of CNN, RNN, LSTM, and Vision Transformer architectures. Hands-on experience with PyTorch and the Hugging Face ecosystem. Understanding of the LLM inference pipeline, tokenization, sampling strategies, and KV caching. Familiarity with fine-tuning techniques (LoRA/QLoRA, SFT, instruction tuning) and experiment tracking. Good knowledge of graph and vector databases and RAG implementation. Strong data structures, algorithms, problem-solving, and analytical skills. Familiarity with version control systems such as Git. Ability to read research papers critically and communicate findings clearly. Good to Have: Exposure to quantization, pruning, and distillation techniques for model compression. Experience with inference runtimes such as llama.cpp, Ollama, vLLM, ONNX Runtime, or TensorRT. Familiarity with agentic frameworks (LangGraph, MCP-based tooling) and multi-step reasoning systems. Exposure to GPU/NPU programming concepts (CUDA, OpenCL) and profiling tools. Exposure to FastAPI / Django / SQLAlchemy for serving prototypes. Publications, preprints, open-source contributions, or competition results in a relevant area. Working knowledge of C/C++.

1) Fill registration form. 2) Attend the quiz online (approx. 30 mins) 3) Schedule coding Test in Electrifex office at Kochi 4) Technical interview 1 in Electrifex office at Kochi 5) Technical interview 2 in Electrifex office at Kochi 6) HR interview in Electrifex office at Kochi