Deep Learning Engineer

2y relevant experience

Not Qualified

For hiring agencies & HR teams

EU engineers, ready to place with your US clients

Pre-screened on AI. Remote B2B contracts. View 5 full profiles free — AI score, skills report, interview questions included.

Executive Summary

This candidate is an enthusiastic data science professional transitioning from a software engineering background, currently completing an MSc at Heriot-Watt University. their ML credentials are real but remain at a foundational-to-intermediate academic level rather than the production-grade deep learning engineering the role demands. Key technical gaps include PyTorch, CUDA/GPU optimization, scalable model deployment, and MLOps tooling — all of which are central to this position. their software engineering background and team leadership experience add value but do not compensate for the ML depth deficit. At this stage of their career, Akarshan would be better suited for a junior ML engineer or data scientist role where growth into production DL is expected over time rather than required from day one.

Top Strengths

✓Genuine ML enthusiasm backed by a relevant MSc degree with an A-grade dissertation
✓Applied NLP experience building an LSTM-based translation system in a professional setting
✓Software engineering fundamentals from 4+ years at Nissan Digital provide a solid coding base
✓Cross-functional experience spanning data science, web development, and backend systems
✓Team leadership experience (led 5 engineers) demonstrates early management readiness

Key Concerns

!Does not meet the production-grade deep learning bar — no PyTorch, no MLOps tooling, no CUDA/GPU optimization, and no deployment experience at scale
!Meaningful employment gap and career transition risk — MSc ongoing through 2025, limited recent hands-on ML work, and primary career was software engineering rather than ML engineering

Culture Fit

58%

Growth Potential

Moderate

Salary Estimate

$55k-$80k (below stated range, given junior-to-mid ML experience level and recent graduate status)

Assessment Reasoning

Akarshan does not meet the minimum threshold for this mid-level Deep Learning Engineer role. they satisfies roughly 40-45% of required skills — Python and TensorFlow/neural network basics are present, but PyTorch (the preferred framework), production MLOps tooling, CUDA/GPU optimization, transformer-based NLP at scale, and Docker/Kubernetes are all absent or undemonstrated. their total experience is 6 years but only ~1-2 years are meaningfully ML-related, and that work was largely academic or small-scale tooling at Nissan. The role requires someone who can independently ship production-grade deep learning systems for candidate matching and NLP inference — a bar Akarshan has not yet reached. This candidate is scored at 48, placing him in the NOT_FIT category. they may be worth revisiting in 1-2 years with more hands-on production ML experience.

Interview Focus Areas

Depth of dissertation work — what models were used, how were they trained, what production considerations were addressed?PyTorch familiarity — any self-study or projects outside TensorFlow?Understanding of transformer architectures beyond textbook — fine-tuning, attention mechanisms, RLHF exposure?How has the candidate approached MLOps or model serving in any context?

Code Review

FairJunior Level

No GitHub or code samples were provided, limiting code quality assessment to resume project descriptions alone. The described projects (LSTM translator, neural network comparison study) are solid academic exercises but do not demonstrate production software engineering maturity. Estimated at junior level based on available evidence — clean, functional code is plausible but optimization, scalability, and software design rigor are unverified.

PythonTensorFlowFlaskNumPyPandasMatplotlibD3.jsJavaSQL

+Evidence of end-to-end project delivery (translator integrated with Flask web app)
+Familiarity with common ML libraries (TensorFlow, NumPy, Pandas, Matplotlib)

-No GitHub profile or code samples were submitted for direct code review
-Project complexity described is modest — LSTM translator and COVID dashboard are standard graduate-level exercises, not production-grade systems
-No evidence of code optimization, test coverage, modular architecture, or engineering best practices beyond basic implementation

Experience Overview

6y total · 2y relevant

This candidate is a recent MSc Data Science graduate with a background primarily in software engineering at Nissan Digital, where they touched ML in a limited capacity (LSTM translation tool). their deep learning experience is largely academic or project-based, and key production-readiness skills (PyTorch, MLOps, CUDA, scalable deployment) are absent or undemonstrated. While they shows genuine interest and foundational knowledge in neural networks, they falls short of the mid-level production DL engineering bar this role requires.

Matching Skills

PythonTensorFlowNeural NetworksLSTM/RNNCNNTransformer (academic)Git/CI-CDPostgreSQLFlaskAWS (basic)

Skills to Verify

PyTorchProduction-grade MLOps (MLflow, Weights & Biases, DVC)CUDA/GPU ComputingVector databasesDocker & KubernetesNLP Transformers at scaleModel inference optimizationComputer Vision at production scale

Candidate information is anonymized. Personal details are hidden for fair evaluation.