-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathcv.tex
225 lines (196 loc) · 14.9 KB
/
cv.tex
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
%% start of file `template.tex'.
%% Copyright 2006-2015 Xavier Danaux ([email protected]).
%
% This work may be distributed and/or modified under the
% conditions of the LaTeX Project Public License version 1.3c,
% available at http://www.latex-project.org/lppl/.
\documentclass[11pt,a4paper,sans]{moderncv} % possible options include font size ('10pt', '11pt' and '12pt'), paper size ('a4paper', 'letterpaper', 'a5paper', 'legalpaper', 'executivepaper' and 'landscape') and font family ('sans' and 'roman')
% moderncv themes
\moderncvstyle{classic} % style options are 'casual' (default), 'classic', 'banking', 'oldstyle' and 'fancy'
\moderncvcolor{blue} % color options 'black', 'blue' (default), 'burgundy', 'green', 'grey', 'orange', 'purple' and 'red'
%\renewcommand{\familydefault}{\sfdefault} % to set the default font; use '\sfdefault' for the default sans serif font, '\rmdefault' for the default roman one, or any tex font name
%\nopagenumbers{} % uncomment to suppress automatic page numbering for CVs longer than one page
% character encoding
%\usepackage[utf8]{inputenc} % if you are not using xelatex ou lualatex, replace by the encoding you are using
%\usepackage{CJKutf8} % if you need to use CJK to typeset your resume in Chinese, Japanese or Korean
% adjust the page margins
\usepackage[scale=0.8]{geometry}
\setlength{\hintscolumnwidth}{0.2\textwidth}
%\setlength{\hintscolumnwidth}{3cm} % if you want to change the width of the column with the dates
%\setlength{\makecvheadnamewidth}{10cm} % for the 'classic' style, if you want to force the width allocated to your name and avoid line breaks. be careful though, the length is normally calculated to avoid any overlap with your personal info; use this at your own typographical risks...
% personal data
\name{Nand}{Dalal}
%\title{Resumé title} % optional, remove / comment the line if not wanted
%\quote{Some quote} % optional, remove / comment the line if not wanted
\email{[email protected]}
\phone[mobile]{+1~(214)~998~5160}
\homepage{https://nanddalal.github.io/}
%\social[github]{nanddalal}
%\extrainfo{additional information} % optional, remove / comment the line if not wanted
% bibliography adjustements (only useful if you make citations in your resume, or print a list of publications using BibTeX)
% to show numerical labels in the bibliography (default is to show no labels)
%\makeatletter\renewcommand*{\bibliographyitemlabel}{\@biblabel{\arabic{enumiv}}}\makeatother
\renewcommand*{\bibliographyitemlabel}{[\arabic{enumiv}]}
% to redefine the bibliography heading string ("Publications")
%\renewcommand{\refname}{Articles}
% bibliography with mutiple entries
%\usepackage{multibib}
%\newcites{book,misc}{{Books},{Others}}
%----------------------------------------------------------------------------------
% content
%----------------------------------------------------------------------------------
\begin{document}
%\begin{CJK*}{UTF8}{gbsn} % to typeset your resume in Chinese using CJK
%----- resume ---------------------------------------------------------
\makecvtitle
%\section{Master thesis}
%\cvitem{title}{\emph{Title}}
%\cvitem{supervisors}{Supervisors}
%\cvitem{description}{Short thesis abstract}
\section{Experience}
\cventry{Dec 2021--Present}{Software Engineer, Machine Learning}{Google}{}{}{
}
%\subsection{Nines}
\cventry{Oct 2019--Dec 2021}{Senior Machine Learning Engineer}{\href{https://sironamedical.com/}{Nines}}{}{}{}
\cventry{Oct 2018--Oct 2019}{Machine Learning Engineer}{}{}{}{
Startup applying machine learning to improve radiology.
\begin{itemize}
\item Built FDA cleared algorithms for \href{https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfPMN/pmn.cfm?ID=K193351}{detecting emergent conditions on CT head scans} and \href{https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfPMN/pmn.cfm?ID=K202990}{measuring lung nodules on CT chest scans}.
\item Iteratively improved computer vision (classification, detection, segmentation) and NLP (classification, summarization) models through evaluation, error analysis, radiologist feedback, and labeling.
\item Designed and developed systems for using data from the Nines teleradiology clinic for analytics and model development in HIPAA compliant environment.
\item Developed multimodal method of using NLP on radiology reports to study generalization properties of image based models.
\item Incorporated ideas from research papers such as hybrid 2.5D detection models for false positive reduction and clustering embeddings to identify hidden stratification.
\item Established patterns and processes within the engineering team for model development, deployment, and monitoring.
\end{itemize}
}
%\subsection{Clarifai}
\cventry{Feb 2017--Sept 2018}{Senior Research Engineer}{\href{https://www.clarifai.com/}{Clarifai}}{}{}{}
\cventry{Feb 2016--Feb 2017}{Research Engineer}{}{}{}{
Startup building a machine learning platform for a variety of computer vision applications.
\begin{itemize}
\item Improved Clarifai's general model (which predicts over 11k concepts) thus improving downstream services which use its embeddings (for transfer learning and visual search).
\item Worked on transfer learning service allowing users to rapidly train image classification models on their own labeled datasets.
\item Developed solutions for customers across many verticals (ear disease classification, aerial imagery detection, fashion snap and search).
\item Transitioned entire company from an internal neural network toolkit to TensorFlow and maintained research toolkit used to train and serve computer vision models.
\item Scaled training infrastructure from single gpu to multi-gpu/node training.
\item Extended serving infrastructure to run in multiple environments (cloud, on-premise, and mobile).
\end{itemize}
}
\cventry{Aug 2014--Dec 2015}{Research Assistant}{University of Michigan}{}{}{
Deep Learning research advised by \href{https://web.eecs.umich.edu/~honglak/}{Professor Honglak Lee}.
\begin{itemize}
\item Applied convolutional autoencoders to non-linear electromagnetic inverse scattering, improving reconstructions and reducing runtimes of through-the-wall imaging.
\item Model improvements: experimented with adaptive learning rates and multi-task learning methods in order to scale up algorithm to harder datasets.
\item System improvements: integrated matconvnet for GPU acceleration in matlab then ported code to Theano for further speedups.
\item This work was presented as \href{https://personal.ntu.edu.sg/acyucel/conf_papers_abs/C33.pdf}{an abstract at IEEE AP-S/URSI 2016} in Puerto Rico.
\item Worked through Professor Lee's advanced tutorials on deep learning algorithms.
\end{itemize}
}
\cventry{Sept 2015--Dec 2015}{Research Assistant}{University of Michigan}{}{}{
Graph Mining research advised by \href{https://web.eecs.umich.edu/~dkoutra/}{Professor Danai Koutra}.
\begin{itemize}
\item Worked on parallelizing and scaling up VoG, a graph summarization algorithm developed by Professor Koutra.
\item Ported over core components of VoG from original matlab codebase to python in order to leverage better libraries for multiprocessing.
\item Profiled and benchmarked algorithm in order to find bottlenecks and then proposed new parallelizable subgraph generation subroutines.
\item Examined the performance of these new methods by running experiments on multiple large graphs and comparing runtime, compression rate, and structure distribution.
\end{itemize}
}
\cventry{Jan 2015--May 2015}{Corporate R\&D Intern}{Qualcomm Research}{}{}{
Deep Learning Research Team
\begin{itemize}
\item Developed a neural network framework for the Adreno GPU.
\item Implemented core neural network internals from scratch in OpenCL.
\item Profiled and benchmarked OpenCL kernels in order to incrementally improve runtime and increase utilization.
\item Worked directly with the GPU architecture team to understand how to improve performance based on the hardware.
\item Presented this work at an internal GPU conference and later integrated into Qualcomm’s heterogeneous neural processing engine, SNPE.
\end{itemize}
}
\cventry{May 2013--Aug 2013}{Data Science Intern}{\href{https://bookshout.com/}{Bookshout}}{}{}{
Bookshout is a social and group e-reading platform.
\begin{itemize}
\item Developed a collaborative filtering based recommendation engine for suggesting books to users.
\end{itemize}
}
\cventry{Jun 2011--Aug 2012}{Research Assistant}{University of Texas at Dallas}{}{}{
High school capstone project advised by \href{https://davidlary.info/}{Professor David Lary} at the Center for Space Sciences.
\begin{itemize}
\item Used neural networks to cross-calibrate $HNO_3$ observations from multiple satellites to quantify periods of strong denitrification in the polar stratospheres.
\item Presented research paper at the Texas Junior Academy of Science at Texas A\&M (2nd place awards in Earth/Space Sciences).
\end{itemize}
}
\section{Education}
\cventry{2012--2015}{B.S.E., Computer Science Engineering}{University of Michigan}{}{}{
\begin{itemize}
\item GPA: 3.673, Summa Cum Laude
\item Relevant coursework: Machine Learning, Graph Mining at Scale, Operating Systems, Database Management Systems, Theoretical Statistics, Matrix Algebra
\item co-re-founded MSAIL (Michigan Student AI Lab): led weekly applied machine learning tutorials for undergraduate students covering classical methods for regression, classification, and clustering culminating in neural networks.
\item Member of HKN (Eta Kappa Nu) and IEEE-ACM
\end{itemize}
}
\cventry{2008--2012}{High School}{Greenhill School}{}{}{
\begin{itemize}
\item GPA: 3.747
\item Kilmer Award for Excellence in Science (top science student)
\item Book awards for most outstanding student: Vector Calculus \& Differential Equations, AP Calculus BC, AP Chemistry, AP Physics C
\item Started science capstone program where students now regularly work with professors from local universities.
\item Member of Math Club, Science Olympiad, and Policy Debate teams.
\end{itemize}
}
%\section{Languages}
%\cvitemwithcomment{Language 1}{Skill level}{Comment}
%\cvitemwithcomment{Language 2}{Skill level}{Comment}
%\cvitemwithcomment{Language 3}{Skill level}{Comment}
%\section{Computer skills}
%\cvdoubleitem{category 1}{XXX, YYY, ZZZ}{category 4}{XXX, YYY, ZZZ}
%\cvdoubleitem{category 2}{XXX, YYY, ZZZ}{category 5}{XXX, YYY, ZZZ}
%\cvdoubleitem{category 3}{XXX, YYY, ZZZ}{category 6}{XXX, YYY, ZZZ}
%\section{Interests}
%\cvitem{hobby 1}{Description}
%\cvitem{hobby 2}{Description}
%\cvitem{hobby 3}{Description}
%\section{Extra 1}
%\cvlistitem{Item 1}
%\cvlistitem{Item 2}
%\cvlistitem{Item 3. This item is particularly long and therefore normally spans over several lines. Did you notice the indentation when the line wraps?}
%\section{Extra 2}
%\cvlistdoubleitem{Item 1}{Item 4}
%\cvlistdoubleitem{Item 2}{Item 5\cite{book1}}
%\cvlistdoubleitem{Item 3}{Item 6. Like item 3 in the single column list before, this item is particularly long to wrap over several lines.}
%\section{References}
%\begin{cvcolumns}
%\cvcolumn{Nines}{\begin{itemize}\item Matthew Lungren\item David Stavens\end{itemize}}
%\cvcolumn{Clarifai}{\begin{itemize}\item Matthew Zeiler\item David Eigen\end{itemize}}
%\cvcolumn{University of Michigan}{\begin{itemize}\item Honglak Lee\end{itemize}}
%\end{cvcolumns}
% Publications from a BibTeX file without multibib
% for numerical labels: \renewcommand{\bibliographyitemlabel}{\@biblabel{\arabic{enumiv}}}% CONSIDER MERGING WITH PREAMBLE PART
% to redefine the heading string ("Publications"): \renewcommand{\refname}{Articles}
%\nocite{*}
%\bibliographystyle{plain}
%\bibliography{publications} % 'publications' is the name of a BibTeX file
% Publications from a BibTeX file using the multibib package
%\section{Publications}
%\nocitebook{book1,book2}
%\bibliographystylebook{plain}
%\bibliographybook{publications} % 'publications' is the name of a BibTeX file
%\nocitemisc{misc1,misc2,misc3}
%\bibliographystylemisc{plain}
%\bibliographymisc{publications} % 'publications' is the name of a BibTeX file
%\clearpage
%%----- letter ---------------------------------------------------------
%% recipient data
%\recipient{Company Recruitment team}{Company, Inc.\\123 somestreet\\some city}
%\date{January 01, 1984}
%\opening{Dear Sir or Madam,}
%\closing{Yours faithfully,}
%\enclosure[Attached]{curriculum vit\ae{}} % use an optional argument to use a string other than "Enclosure", or redefine \enclname
%\makelettertitle
%Lorem ipsum dolor sit amet, consectetur adipiscing elit. Duis ullamcorper neque sit amet lectus facilisis sed luctus nisl iaculis. Vivamus at neque arcu, sed tempor quam. Curabitur pharetra tincidunt tincidunt. Morbi volutpat feugiat mauris, quis tempor neque vehicula volutpat. Duis tristique justo vel massa fermentum accumsan. Mauris ante elit, feugiat vestibulum tempor eget, eleifend ac ipsum. Donec scelerisque lobortis ipsum eu vestibulum. Pellentesque vel massa at felis accumsan rhoncus.
%Suspendisse commodo, massa eu congue tincidunt, elit mauris pellentesque orci, cursus tempor odio nisl euismod augue. Aliquam adipiscing nibh ut odio sodales et pulvinar tortor laoreet. Mauris a accumsan ligula. Class aptent taciti sociosqu ad litora torquent per conubia nostra, per inceptos himenaeos. Suspendisse vulputate sem vehicula ipsum varius nec tempus dui dapibus. Phasellus et est urna, ut auctor erat. Sed tincidunt odio id odio aliquam mattis. Donec sapien nulla, feugiat eget adipiscing sit amet, lacinia ut dolor. Phasellus tincidunt, leo a fringilla consectetur, felis diam aliquam urna, vitae aliquet lectus orci nec velit. Vivamus dapibus varius blandit.
%Duis sit amet magna ante, at sodales diam. Aenean consectetur porta risus et sagittis. Ut interdum, enim varius pellentesque tincidunt, magna libero sodales tortor, ut fermentum nunc metus a ante. Vivamus odio leo, tincidunt eu luctus ut, sollicitudin sit amet metus. Nunc sed orci lectus. Ut sodales magna sed velit volutpat sit amet pulvinar diam venenatis.
%Albert Einstein discovered that $e=mc^2$ in 1905.
%\[ e=\lim_{n \to \infty} \left(1+\frac{1}{n}\right)^n \]
%\makeletterclosing
%\clearpage\end{CJK*} % if you are typesetting your resume in Chinese using CJK; the \clearpage is required for fancyhdr to work correctly with CJK, though it kills the page numbering by making \lastpage undefined
\end{document}
%% end of file `template.tex'.