@Mary #64: I agree with most of what you say here and what you said before. But: Clearly, srg needs to improve her German and to gain specialised knowledge in a few fields. These skills could be gained, for instance, by a working holiday in a German-speaking country and work experience in one or more of the fields in which she wishes to specialise. I guess the problem is that srg doesn't know in advance in which fields she should specialize, and once she gets an assignment it's too late to specialize.
I'm also afraid that most of us (including myself) do entertain wrong ideas about how the translation biz works in countries like India (or in globalized companies, for that matter). Still, it's unbelievable that somebody should expect 90-100 pages of text within 4 days. I agree with what Res-can said in #5 above. You simply can't do it, physically.
The problem good command of the language(s) involved vs. competence in the field is tricky. If you can't get both, which do you prefer? My experience: You can't translate anything if you don't know what you are talking about. Beautiful English with all the wrong words is certainly less acceptable than poor English with the right terms. But even when you know what you are talking about, it may be difficult to get the meaning across. A grammatical mistake, ambiguous syntax, a wrong declension may mess up a technical translation.
I have done (technical) translations both ways, G/E and E/G, and I have encountered the same problem in both directions. Do I understand what the authors want to say?
My suggestions are:
a) talk to the people who wrote it. "Can you explain to me what you actually do when you compare cumulative frequencies? -- It is not clear to me if you use standard deviation in diagram X. -- Do you mean special 'Schmierpunkte' or the whole of component X when you say 'apply oil'?
b) Get a special dictionary, so you can venture an educated guess.
c) Have somebody read over it if you translate from your first into your second language (and, in srg's case, maybe both). You will soon see where your weaknesses are, and if some native speaker understands your translation you will gain so much confidence that after a few trials you know pretty well where you stand and if you can risk it on your own (and save time or money).
d) If you try to make a living of it, be aware that freelance means lonely. You can avoid this by building up a network, a few people who know the field very well, a few native speakers, anybody who might help you take the initial hurdles. Within a fairly short time you will be able to help others.
e) Read plenty of comparable texts in both languages, but beware of translations. You'll get ideas without knowing where you got them, and they will usually be good ideas.
There are some big companies (Brother is an example) who spend much time on good translations, you can actually understand their manuals in German, and there are others who don't bother :-), but I wonder if perfect German documentation as in the case of Brother is not one little reason for the fact that this company is big. I don't know what the Indian market is like, but if you're not into translating Shakespeare you should always ask yourself if your translated text is useful for the native speaker who reads it. If he or she can handle the product (for example), it's OK even if there are a few mistakes. If not, it was not worth doing.
I'm aware that this strategy (in Germany) means that you earn very little money for a period of several years. This could be the basis for better-paid, more demanding jobs in the future. If you want to do hack work, you'll probably be able to sustain yourself for some time, but it doesn't get you (and nobody else, for that matter) anywhere. It is very difficult to decide, it depends on your present situation and on your personal options and plans for the future.
I hope this could add a few aspects to all the valuable things others have said.